gene expression profiles uncover individual identities of …...which lie in close vicinity to...

STEM CELLS AND REGENERATION TECHNIQUES AND RESOURCES ARTICLE

Gene expression profiles uncover individual identities of gnathalneuroblasts and serial homologies in the embryonic CNS ofDrosophilaRolf Urbach*,‡, David Jussen* and Gerhard M. Technau

ABSTRACTThe numbers and types of progeny cells generated by neural stemcells in the developing CNS are adapted to its region-specificfunctional requirements. In Drosophila, segmental units of the CNSdevelop from well-defined patterns of neuroblasts. Here weconstructed comprehensive neuroblast maps for the three gnathalhead segments. Based on the spatiotemporal pattern of neuroblastformation and the expression profiles of 46 marker genes (41transcription factors), each neuroblast can be uniquely identified.Compared with the thoracic ground state, neuroblast numbers areprogressively reduced in labial, maxillary and mandibular segmentsdue to smaller sizes of neuroectodermal anlagen and, partially, tosuppression of neuroblast formation and induction of programmedcell death by the Hox geneDeformed. Neuroblast patterns are furtherinfluenced by segmental modifications in dorsoventral and proneuralgene expression. With the previously published neuroblast maps andthose presented here for the gnathal region, all neuroectodermalneuroblasts building the CNS of the fly (ventral nerve cord and brain,except optic lobes) are now individually identified (in total 2×567neuroblasts). This allows, for the first time, a comparison of thecharacteristics of segmental populations of stem cells and to screenfor serially homologous neuroblasts throughout the CNS. We showthat approximately half of the deutocerebral and all of the tritocerebral(posterior brain) and gnathal neuroblasts, but none of theprotocerebral (anterior brain) neuroblasts, display serial homologyto neuroblasts in thoracic/abdominal neuromeres. Modifications in themolecular signature of serially homologous neuroblasts are likely todetermine the segment-specific characteristics of their lineages.

KEY WORDS: Central nervous system, Neuroblasts, Segmentalpatterning, Drosophila brain, Gene expression profile, Deformed

INTRODUCTIONThe development of the central nervous system (CNS) inDrosophila begins with the formation of a stereotyped populationof neural stem cells, termed neuroblasts (NBs), which delaminatefrom the neuroectoderm in a precise spatiotemporal pattern.Positional cues within the neuroectoderm provided by products ofearly regulatory genes control the identity of embryonic NBs(e.g. reviewed by Skeath and Thor, 2003). Each NB within a

hemisegment acquires a unique identity that is reflected in thetypical developmental time point and position of its delaminationfrom the neuroectoderm, the combinatorial code of genes itexpresses (Hartenstein and Campos-Ortega, 1984; Doe, 1992;Urbach and Technau, 2003b), and the production of a specific celllineage (Bossing et al., 1996; Schmidt et al., 1997; Schmid et al.,1999). The developmental patterns and identities of embryonic NBshave been described in detailed maps for the brain (Younossi-Hartenstein et al., 1996; Urbach et al., 2003; Urbach and Technau,2003a,b) and the thoracic (T1-T3) and anterior abdominalneuromeres (A1-A7) of the ventral nerve cord (VNC) (Doe, 1992;Broadus et al., 1995). In contrast to the brain, neuromeres T1-A7originate from a rather stereotypic array of ∼30 NBs perhemisegment. In each hemisegment, the Cartesian grid-likeexpression of anteroposterior (AP) and dorsoventral (DV)patterning genes is virtually identical. Hence, NBs developingfrom the same ‘quadrants’ in different segments represent serialhomologs, as they acquire a similar identity (reviewed by Skeath andThor, 2003; Technau et al., 2006). Recently, NB patterns were alsoestablished for the derived terminal abdominal neuromeres (A8-A10), and were shown to consist of segment-specifically reducedsets of serial homologous NBs (Birkholz et al., 2013a).Accordingly, the gnathal neuromeres, which constitute thesubesophageal zone (between brain and truncal neuromeres)comprise the last region of the fly CNS in which patterns andidentities of NBs have not been resolved so far. The labial (LB),maxillary (MX) and mandibular (MN) neuromeres (from posteriorto anterior) develop in the gnathal part of the head, in whichsegmental units are more evident compared with the pregnathalhead from which the embryonic brain arises (Schmidt-Ott andTechnau, 1992; Urbach and Technau, 2003a). During ongoingdevelopment, these neuromeres fuse to become the subesophagealzone at the posteriormost site of the adult brain (Ito et al., 2014), aregion implicated in feeding and the taste response (e.g. Scott et al.,2001; Gendre et al., 2004; Melcher and Pankratz, 2005).

In this study, we have undertaken a comprehensive survey ofthe early development of the gnathal neuromeres. We traced thespatiotemporal pattern of formation of the gnathal NBs. Within thefinal NB pattern by stage 11, numbers are progressively diminishedin LB, MX and MN (i.e. in posterior-anterior order), as comparedwith thoracic neuromeres. This is mainly due to smaller segmentalsizes of gnathal neuroectodermal anlagen, modifications inpatterning gene expression, and by the activity of Deformed(Dfd), which supresses NB formation. Moreover, we providecomprehensive maps of the ∼76 gnathal NBs (within the threehemisegments, plus three unpaired midline NBs), which reveal theexpression patterns of 46 different marker genes (encoding 41transcription factors), that are specifically expressed in particularNB subsets and form combinatorial codes specifying each NBReceived 2 December 2015; Accepted 22 February 2016

Institute of Genetics, University of Mainz, Mainz D-55099, Germany.*These authors contributed equally to this work

‡Author for correspondence ([email protected])

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

1290

© 2016. Published by The Company of Biologists Ltd | Development (2016) 143, 1290-1301 doi:10.1242/dev.133546

DEVELO

PM

ENT

mailto:[email protected]://creativecommons.org/licenses/by/3.0http://creativecommons.org/licenses/by/3.0

individually. Thus, detailed spatiotemporal and molecular mapsnow exist for the entire population of NBs giving rise to the brain(except for the optic lobes) and VNC of the fly (in total 2×567 NBsof neuroectodermal origin). The completed map makes it possible,for the first time, to compare the patterns and molecularcharacteristics of segmental populations of NBs throughout theCNS and to identify serial homologies. Our data demonstrate thatalmost all gnathal NBs are serially homologous to NBs in moreposterior segments, and provide support that most NBs in thetritocerebrum, and about half of the NBs in the deutocerebrum,show serial homology to NBs in the VNC.This study provides a basis for investigating, at the level of

identified NBs and their lineages, the mechanisms that underlie thestructural and functional diversification of the segmental CNS units.It will also facilitate comparisons of the patterns and molecularprofiles of neural stem cells among different species in the contextof evolutionary investigations.

RESULTSSpatiotemporal pattern of NB formation in the gnathal headsegmentsWe traced the pattern ofNBs in the gnathal segments in comparison toT1 in flat preparations of fixed Drosophila embryos during stages 8-12. This developmental periodwas subdivided into six stages,most ofwhich match those previously reported in the truncal CNS(Hartenstein et al., 1987; Doe, 1992; Broadus et al., 1995; Birkholzet al., 2013a) and brain (Urbach et al., 2003). NBs were identified bysize, subectodermal position and expression of the stem cell markersdeadpan (dpn) and worniu (wor). The final NB pattern is establishedby late stage 11. At that stage, dpn and wor are expressed not only inNBs but also in gnathal sensory organ precursors (SOPs), some ofwhich lie in close vicinity to dorsal NBs. The SOP-specific markercousin of atonal (cato) (Goulding et al., 2000) allowed us tounambiguously discriminate NBs from SOPs (Fig. S1). By means ofthe stereotypic position and time at which each NB develops withinthe neuroectoderm, and expression of the five marker genes engrailed(en; indicating posterior NBs and segmental boundaries) (DiNardoet al., 1985; Doe, 1992), intermediate neuroblasts defective (ind;indicating intermediate NBs) (Weiss et al., 1998), odd skipped (odd),seven up (svp)-lacZ (Broadus et al., 1995) and empty spiracles (ems)(all expressed in specific NB subsets) we were able to identifyindividual gnathal NBs in the developing NB pattern (Fig. 1). Thespatiotemporal pattern of gnathal NB development is largely invariant(among specimen). Whereas the formation of NBs in LB and MXresembles that of truncal segments, it is significantly delayed in MN,particularly in anterior positions. As compared with T1 (32 NBs perhemisegment), the final number of NBs is reduced in all gnathalsegments, decreasing from LB to MN: aside the unpaired median NB(MNB), we found ∼28 NBs in LB, ∼26 NBs in MX, and 22 NBs inMN per hemisegment (Fig. 1F).

Each gnathal NB expresses a specific combinatorial code ofmarker genesIn order to further characterize individual gnathal NBs in our mapwe investigated the expression of 64 NB marker genes (usingantibodies, in situ probes or lacZ lines; see Table S1), most of whichhave previously been used to map NBs in the thorax, abdomen andbrain (Doe, 1992; Broadus et al., 1995; Younossi-Hartenstein et al.,1996; Urbach and Technau, 2003b; Birkholz et al., 2013a).Eighteen of these marker genes, expressed in subsets of brainNBs (Urbach and Technau, 2003b; Urbach, 2007; Kunz et al., 2012;data not shown), were not expressed in gnathal (or thoracic) NBs

(Table S1). Thus, we identified 46 marker genes as expressed in all[asense (ase), wor, dpn] or in specific subsets of gnathal NBs(including two marker genes for typical lineage components) asshown in detail in Fig. 2 (see also Figs S2.1-S2.6) and summarizedin Fig. 3. Among these we provide novel markers for NB subsets:knirps (kni), buttonhead (btd), Dbx, Centaurin gamma 1A(CenG1A), charybde (chrb), collier (col; knot – FlyBase) andgiant (gt). All of the ∼76 NBs of the three gnathal hemisegments(plus three unpaired MNBs) are individually identifiable by theircharacteristic developmental time point, neuroectodermal origin(along AP and DV axes) (Fig. 1, Table S2) and by their uniqueexpression profiles (Figs 2 and 3, Figs S2.1-S2.6).

Serial homology of gnathal and thoracic/abdominal NBsAs 32 of the 46 marker genes (indicated with an asterisk in Table S1)are expressed in segmentally repeated subsets of thoracic andabdominal NBs, we used the expression of these ‘segmentallyconserved’marker genes as indicators for serial homology of gnathalNBs, in addition to the position and time point of their formation(Fig. 1,Table S2). ComparedwithT1 (resembling the ground state)wefound these characteristics to be strongly conserved in correspondingNBs in LB and MX and, despite more pronounced modifications,largely in the MN. Only in a few cases was the serial homology ofmandibularNBs ambiguous. For example,NB5-5 (lacking inMXandLB) expresses the typical markers gooseberry (gsb)-lacZ, wingless(wg) and sloppy paired 1 (slp1), but lacks expression of unplugged(unpg)-lacZ, huckebein (hkb)-lacZ and svp-lacZ, and atypicallyexpresses chrb, Gt, Dachshund (Dac) and Ladybird early (Lbe)(Figs 2 and 3, Figs S2.1, S2.2 and S2.4). Further, we identified ∼3NBs per mandibular hemisegment that express an MP2-like markergene profile of Prospero (Pros), Achaete (Ac), Hunchback (Hb) andVentral nervous system defective (Vnd), but lack the MP2-specificmarkers Odd and Aj96-lacZ (lacZAJ96) (Fig. 3C-E, Fig. S2.6A-C).Pros is nuclear from their time of formation and is not localizedcortically during the largely symmetric division (Fig. S2.6D,E), as ischaracteristic for MP2 (Spana and Doe, 1995). Considering theirmolecular profile and division behavior and the fact that they allemerge from the same ac/scute (sc)-coexpressing proneural domain(see below), they might represent ‘duplicated’ serial homologs of theMP2s found in truncal segments.

Lack of marker gene expression indicated the segment-specificabsence of particular gnathal NBs (compared with T1), as shown forsome examples in the following (Fig. 3, Fig. S2 and as detailed inTable S2). Lack of the expression profile Dpn+/Pox neuro (Poxn)+/Eagle (Eg)+/hkb-lacZ+ revealed the absence of NB2-4, and lack ofthe Dpn+/mirror (mirr)-lacZ+/En–/Runt (Run)+ profile revealed theabsence of NB2-3, both in all gnathal segments. Lack of the Dpn+/Eyeless (Ey)+/Ind+/Dbx+/Run+ profile demonstrated loss of NB3-2,and lack of Dpn+/Eg+/Ems+/svp-lacZ–/Run+ loss of NB3-3, bothspecifically in MN. Our data show that segmental modifications inthe NB pattern are due to dorsal NBs missing preferentially in theanterior compartment of the gnathal segments. Accordingly, NBs 2-3, 2-4 and 5-5 are lacking in LB, and NBs 2-3, 2-4, 2-5, 5-4, 5-5 andMP2 are lacking inMX. NBs 1-3, 2-3, 2-4, 2-5, 3-2, 3-3, 5-4 and thelongitudinal glioblast (LGB) are missing in MN, but so are theventral NBs 1-1, 1-2, 5-1 and, exceptionally, the posterior NB6-4.There seems to be no correlation between NBs lacking in gnathalsegments and their time point of formation in T1.

In summary, although MD shows the most significant reductionin NB numbers we identified a potential NB5-5 (missing in MXand LB) and ∼3 MP2-like NBs instead of one in the otherhemineuromeres (except MX, where MP2 is missing). Based on

1291

STEM CELLS AND REGENERATION Development (2016) 143, 1290-1301 doi:10.1242/dev.133546

DEVELO

PM

ENT

http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1

their individual expression profiles, all NBs in LB and MX andalmost all NBs in MN have serially homologous counterparts in thethoracic/abdominal segments.

Gene expression profiles in serially homologous NBs areprogressively modified from LB to MNNext, we estimated the extent to which gene expression is modifiedin individual gnathal NBs by considering those 32 ‘conserved’marker genes (indicated with an asterisk in Table S1) that areexpressed in segmentally repeated subsets of thoracic/abdominalNBs (Table S2). We often observed that, compared with T1,expression of genes is lacking or they are ‘ectopically’ expressed: intotal, the expression of eight genes was altered in a subset of 11 (outof 29) NBs in LB, the expression of ten genes in a subset of 14 (outof 26) NBs in MX, and the expression of 18 genes in a subset of 17

(out of 22) NBs in MN (Fig. 3F). This indicates that profiles ofsegmentally expressed genes in serially homologous NBsprogressively differ from LB to MN. We identified further genesto be exclusively (col, gt, chrb) or preferentially (dac, CenG1A)expressed in particular subsets of mandibular NBs (Fig. 2E,F,Fig. 3E, Figs S2.1 and S2.4). Considering these factors altogether,the expression profile of almost all mandibular NBs is alteredcompared with corresponding NBs in other VNC neuromeres. Thus,the identity of many gnathal, and in particular mandibular, NBs hasundergone segmental modifications.

Identification of serially homologous NBs in the neuromeresof trunk and brainWe further investigated the extent to which individual NBs in thebrain and trunk share specific developmental and molecular

Fig. 1. Spatiotemporal development of the NB pattern in the gnathal segments. (A-F) Semi-schematic representations of ventral views of the left half of thegnathal segments (MN, mandibular; MX, maxillary; LB, labial) and the prothoracic segment (T1), as framed in the inset in A. Typical NB arrangement is shown at(A) late stage 8 (lst8), (B) stage 9 (st9), (C) early stage 10 (est10), (D) early stage 11 (est11), (E) late stage 11 (lst11) and (F) early stage 12 (est12). The position,formation time point and expression of five marker genes (see color code) allows the identification of individual NBs. NB formation is significantly delayed inMN, similar to observations made in tritocerebrum (TC; NBs light gray) (Urbach et al., 2003). Existence of NB1-3 is unclear in MX and LB (see E and Fig. S2.5).(F) Dorsal labial NBs become separated by salivary gland anlagen (SGA). is, intercalary en stripe; MNB,median neuroblast; ML, ventral midline; LGB, longitudinalglioblast; TA, tracheal anlagen (see also following figures).

1292


DEVELO

PM

ENT

http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1

traits, which may support serial homology (Urbach and Technau,2004). Although the expression of many marker genes hasalready been described in brain NBs (Urbach and Technau,2003b), further markers were analyzed in brain NBs that aresegmentally expressed in truncal NBs [Nkx6 (HGTX – FlyBase),Dbx, H15-lacZ, Midline (Mid), btd, Ind] (Table S2; and data notshown) to allow a comprehensive comparison of their molecularsignatures. When comparing the developmental time point,neuroectodermal origin and specific molecular signature of

individual thoracic and gnathal NBs with those in the twoposterior brain neuromeres, i.e. the tritocerebrum anddeutocerebrum, remarkable parallels were observed(summarized in Fig. 4 and Table S2). Accordingly, we couldattribute ten out of 13 NBs in the tritocerebrum to correspondingNBs in the thoracic and, usually, gnathal neuromeres. Each of theremaining three NBs (Td3, Tv4,5) shares features with two orthree thoracic NBs, but could not be unambiguously assigned(see Table S2). Although all NBs in the tritocerebrum develop

Fig. 2. Mapping and identificationof NBs in gnathal neuromeres.(A-G) Composite confocal images offlat preparations (ventral view) of latestage 11 embryos stained for differentcombinations of molecular markers asindicated. Subsets of NBs identifiedby marker staining(s) and position arelabeled. (A′,A″,B′,B″,C′,C″,D′,D″,G′,G″) Left side. (A-A″) Ind+ NB3-2 islacking in MN. (B-B″) Lbe is atypicallyexpressed in mandibular NB5-5.(C-C″) Run is found in increasinglysmaller NB subsets from LB toMN. Eyis expressed in a reduced NB subsetin MN. (D-D″) mirr-lacZ is detected inreduced NB subsets in gnathalsegments, and Ems in MN. (E,E′) Colis exclusively expressed in the fouranteriormost mandibular NBs/hemisegment. (F,F′) Dac isexclusively expressed inMN, in four tofive anterior NBs/hemisegment.(G-G″) Mid is expressed in a reducedNB subset in MN. Repo+ glial cellsderive from NB6-4, 7-4 and LGB.

1293


DEVELO

PM

ENT

http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1

from neuroectodermal positions similar to those of their truncalhomologs, most of them develop significantly later, comparableto the corresponding NBs in MN.

Applying these criteria to individual NBs in the deutocerebrum,∼9 out of 21 NBs exhibit strong parallels with specific NBs in thethoracic and gnathal neuromeres. Usually, the gene expression

Fig. 3. NB maps summarizing the expression of AP and DV patterning genes, temporal genes and other NB marker genes in gnathal neuromerescompared with the prothoracic neuromere. (A-E) Expression of subgroups of marker genes (as indicated by color code) in the full complement of gnathal NBs(left side) at late stage 11; sublineage markers for specific NBs in D are indicated by outer circles. (F) Gray columns represent the total number of NBs in therespective segments. Colored columns indicate numbers of NBs in which expression of the indicated genes is ‘ectopically′ detected (red) or lacking (blue),compared with T1. For example, in MN 14 NBs exhibit ‘ectopic’ expression of one or more of those genes (out of the ten indicated), whereas 13 NBs lackexpression of one or more other genes (out of the 11 indicated). In some NBs, expression of certain genes is ‘ectopically’ detected and that of others is lacking(indicated by overlapping columns).

1294


DEVELO

PM

ENT

profiles of deutocerebral NBs show greater similarity to those ofcorresponding NBs in truncal neuromeres than to those in thetritocerebrum. In some cases, strong parallels were obvious betweenanNB in the deutocerebrum (Dd3,4,6) and in the truncal neuromeres,but a corresponding NB was missing in the tritocerebrum.Conversely, for some NBs in the tritocerebrum (Td5,7, Tv3) thatshare similarities with specific NBs in the trunk, corresponding NBscould not be identified in the deutocerebrum (Fig. 4).These data demonstrate close correspondence between individual

NBs in the posterior brain (trito- and deutocerebrum) and trunk,supporting the assumption that these NBs are serially homologous.NBs inMX represent a subset of those in LB, and those inMN largelyrepresent a subset of those in MX. Interestingly, NBs in thetritocerebrum again seem to represent a subset of NBs in MN.Although our data support the existence of serially homologous NBsalso in the deutocerebrum, less than half of the deutocerebralNBs andnone of the ∼70 protocerebral NBs appear to have counterparts inmore posterior segments. Thus, neuromeres following T1 anteriorlyexhibit an increasing degree of derivation from the ground state.

Gnathal NB4-2 lineages reveal segment-specificmodifications with regard to RP2 developmentSegmental modifications in gene expression profiles of gnathal NBsare reflected by their cell lineages, as judged from the expression of

sublineage markers [e.g. Even skipped (Eve), eveRRK-Gal4, CQ-Gal4] (Fig. 3D, Fig. S2.5). Eve and eveRRK-Gal4 are continuouslycoexpressed in the NB4-2-derived RP2 motoneuron in thoracic/abdominal segments (Bossing et al., 1996). InMN, however, we didnot detect Eve/eveRRK>GFP+ cells at any stage, indicating thatmandibular NB4-2 does not form RP2. By contrast, in LB and MX,we observed that RP2 is often lacking at stage 15, althoughconsistently present at earlier stages (Fig. S2.5). In cell death-deficient Df(3L)H99 embryos (White et al., 1994), Eve+ RP2 isrestored to 100% in MX and LB at stage 15 (n=22 hemisegments),indicating that RP2 normally undergoes programmed cell death(PCD) in both segments (Fig. S2.5, Table S3). Thus, the gnathalNB4-2 lineages reveal segment-specific diversification of RP2,which does not develop in MN and undergoes PCD in MX and LB.

PCD does not significantly account for the reduction in NBnumbers in gnathal segmentsTo investigate whether PCD regulates segment-specific differencesin the gnathal NB pattern we performed antibody staining againstDeath caspase-1 (Dcp-1), an early hallmark of cell death. During theperiod of NB formation, until early stage 12, Dcp-1 signal waslargely absent from the labial neuroectoderm, and detectedparticularly in neuroectoderm of the posterior mandibular andanterior maxillary compartment; it was not detectable in gnathalNBs (Fig. 5K). This suggests that PCD does not account for thereduced NB numbers in LB.

To investigatewhetherNB formation is affected byPCDoccurringin the posteriorMN and anteriorMX,we analyzed theNB patterns inDf(3L)H99 embryos at late stage 11. The spatial arrangement andtotal number of Dpn+ NBs in all mutant gnathal segments did notobviously differ from wild type (Fig. 5H). Nevertheless, using morespecific markers, in 15% of hemisegments (n=52) an ectopic Eg/En-coexpressing NB was observed at the position of NB6-4 in mutantMN (Fig. 5N,N′). OtherNBswere never restored in gnathal segmentsofDf(3L)H99mutants. These data suggest that, with the exception ofmandibular NB6-4, PCD does not significantly account for thereduction in NB numbers in gnathal segments.

Deformed suppresses the formation of specific NBs ingnathal segmentsBecause the homeotic gene Deformed (Dfd) is expressed in MN andMX (Fig. 5I,J) (McGinnes et al., 1998), we tested its role in shapingthe gnathal NB pattern. InDfd16mutant embryos, we found a slightlyincreased number ofDpn+NBs inMX (Fig. 5H).Usingmore specificNB markers (e.g. Eg, Ey, En, Ind, Lbe, Vnd, Wg) we often detectedan ectopic (Wg+) NB at the position of NB5-4 (81% ofhemisegments, n=26; Fig. 5E-G′) and, in one case (n=48hemisegments), an ectopic Eg+ NB at the position of NB2-4(Fig. 5C) in the mutant MX. Furthermore, in the mutant MN, anectopic Eg/En-coexpressing NB at the position of NB6-4 was foundin 20% of hemisegments (n=40; Fig. 5A,B). However, NBs werenever restored in the anterior mandibular compartment, where, in thewild type, the reduction of theNBpattern ismost substantial. Thus, inthis region Dfd does not seem to play a role in suppressing NBformation, in accordance with our observation that during the periodof NB formation Dfd is already largely downregulated in theneuroectoderm of the anterior mandibular compartment (Fig. 5I). Inrare cases, an ectopic Eg+ NB at the position of NB2-4 was alsodetected in LBofDfd16mutants (2/48 hemisegments; Fig. 5D). Sincein wild type we detected Dfd protein in the neuroectoderm fromwhich labial NB2-4 seems to develop (Fig. 5I, inset), this suggeststhat Dfd acts cell-autonomously to suppress labial NB2-4 formation.

Fig. 4. Serial homologous NBs in neuromeres of the trunk and brain.Assignment of potentially serially homologous NBs (same color) inneuromeres of brain (TC, tritocerebrum; DC, deutocerebrum; PC,protocerebrum), gnathal segments (MN, mandibular; MX, maxillary; LB, labial)and prothorax (T1; resembling the ground state), based on developmental timepoint, neuroectodermal origin (in AP and DV axes) and the specific molecularsignature of individual NBs (as summarized in Table S2). A few NBs in TC andDC show serial homology to two NBs in neuromeres of ventral nerve cord(VNC). The brain NB map is according to Urbach et al. (2003).

1295


DEVELO

PM

ENT

http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1

As we observed an ectopic mandibular NB6-4 both inDf(3L)H99andDfd16mutants, we tested inDfd16mutants whether this is due toa decrease in PCD in the mandibular neuroectoderm.We observed asubstantially reduced number of Dcp-1+ cells in MN and MXneuroectoderm (Fig. 5K-M), including the En-expressing dorsalneuroectoderm from which the ectopic mandibular NB6-4develops: in wild type, Dcp-1 was detected in 31% ofhemisegments (n=64), versus 8% of hemisegments (n=26) inDfd16 mutants (Fig. 5K, inset). Thus, Dfd positively regulates PCDin neuroectodermal progenitor cells and, thereby, suppresses

formation of the mandibular NB6-4. By contrast, the maxillaryNB5-4, which was often restored in Dfd16 mutants (see above), wasnever found in Df(3L)H99 mutants (n=24 hemisegments). Thesefindings suggest that Dfd does not exclusively act via PCD to restrictNB formation.

The number of neuroectodermal progenitors differs betweengnathal segmentsSince PCD does not significantly contribute to segment-specificrestriction of NB numbers in the gnathal segments, we next

Fig. 5. Roles of Dfd, programmed cell death and the number of neuroectodermal progenitors in regulating NB numbers in gnathal segments.(A-D) Gnathal (MN, MX, LB) and prothoracic (T1) left hemisegments in wild type (wt) (A) or Dfd16 mutants (B-D). White dotted lines outline ectopic Eg+ NBs.(E,F) Lbe+ and Ey+ NBs are indicated. Note the ectopic Dpn+ NB5-4-like cell (e5-4) in F. (G,G′) Ectopic NB5-4 expresses row 5-specific Wg (G) and is positioneddorsally to Ind+ NB5-3 (G′). (H) Number of Dpn+ NBs in gnathal and prothoracic hemisegments in wild type (MN, 20.8±1.5; MX, 24.6±1.0; LB, 26.9±0.7; T1, 28.8±0.5; n=14 each),Df(3L)H99 (H99; MN, 21.2±1.1; MX, 25.7±1.4; LB, 27.5±2.0; T1, 29.2±1.4; n=17 each) andDfd16 (MN, 21.9±0.9; MX, 27.8±1.8; n=16 each). ND,not determined (Dfd expression largely restricted to MN and MX). (I,J) Dfd expression in outer neuroectoderm (I) and within the NB layer (J). (I) Note thelow level of Dfd in the neuroectoderm of MN compared with MX. Inset shows Dfd in anterodorsal labial neuroectodermal cells (arrow). (K-N′) At early stage12 Dcp-1 signal is reduced in Dfd16 mutant mandibular and maxillary neuroectoderm. Solid white lines indicate the dorsal border of neuromeres. (M) Theaverage number of Dcp-1+ neuroectodermal (NE) cells is diminished in Dfd16 MN and MX. wt, late stage 11 (lst11): MN, 2.6±1.8; MX, 5.1±3.1 (n=14 each).Dfd16 lst11: MN, 0.4±1.6; MX, 0.8±1.2 (n=28 each). wt, early stage 12 (est12): MN, 7.0±5.1; MX, 15.5±5.7 (n=29 each). Dfd16 est12: MN, 2.5±2.0; MX, 5.5±4.5(n=32 each). (N,N′) Note the ectopicmandibular NB6-4 (e6-4) inDf(3L)H99. (O) Stage 9. Red lines indicate segmental borders, blue lines dorsal neuroectodermalborder, black line the ventral midline. In LB, four spatial quadrants are indicated: ad, anterodorsal; av, anteroventral; pd, posterodorsal; pv, posteroventral.(P) Neuroectodermal cell numbers in gnathal and prothoracic hemisegments and in each quadrant of each hemisegment (color coded). n=7-11 (all quadrants).(A-G′,I-L,N,N′) Composite confocal images. (H,M,P) Data are mean±s.d. **P

examined whether it is related to the number of neuroectodermalprogenitors from which those NBs develop. Upon staining againstMsh (Drop – FlyBase) and En, neuroectodermal cells werecounted in all four spatial quadrants of each gnathal andprothoracic hemineuromere at early stage 9, when expression ofEn (in posterior cells) and Msh (in dorsal cells) is sufficientlyestablished, and before onset of PCD (Fig. 5O). Compared withT1, the total cell number was reduced in MX (by ∼15%) and MN(by ∼43%), but not in LB. Whereas in MX cell numbers werediminished only in the anterior quadrants, in MN a reduction wasfound in all four quadrants, but most significantly in theanterodorsal quadrant (by ∼64%; Fig. 5P). These findingsmatch the NB pattern of both segments, where NBspreferentially in anterodorsal positions do not form. Therefore,early determination of smaller segmental sizes of gnathalneuroectodermal anlagen appears to mainly define the numbersof delaminating NBs. The various regions within segmentalneuroectodermal anlagen (e.g. the anterior compartment of MN)are differentially affected by size reduction.

Modifications in DV gene expression establish an expandeddomain of proneural gene expression to allow for rapidformation of NBs in the MNAlthough NB formation is completed by late stage 11, we find thatmany NBs of the MN (∼60%), particularly in the anteriorcompartment, are delayed compared with the other gnathalsegments (Fig. 1). Accordingly, most mandibular NBs developin a fairly narrow time window and, as shown above, from acomparatively small neuroectoderm. To see if this peculiarity isreflected in the activity of proneural genes we investigated theexpression of ac, sc and lethal of scute (l’sc; l(1)sc – FlyBase) inthe neuroectoderm. The early spatiotemporal expression of thesegenes in LB, MX and the posterior compartment of MN is similarto the situation in truncal segments, where each NB develops froma small proneural cluster of neuroectodermal cells (Fig. 6F-K)(Jiménez and Campos-Ortega, 1990). In the anterior compartmentof MN, however, proneural gene expression starts slightly later, inaccordance with the postponed development of many anterior NBs(Fig. 1). Strikingly, during stages 9 to 11, ac and sc arecontinuously expressed in this compartment at high levels in anoversized, neuroectodermal stripe (Fig. 6F-I), from which (bystages 10/11) eight to nine NBs develop in a narrow time window;these include part of row 3, 4, 5 NBs and the MP2-like NBs(Fig. 6A-C). Expression of l’sc begins slightly later than that ofac/sc and, until stage 11, when it is already downregulated in MXand LB, it is expressed within a large domain covering the MN(Fig. 6J,K). Unlike ac and sc, l’sc is also expressed in the mostanterior neuroectoderm of MN, suggesting that development of themost anterior NBs (i.e. row 2) depends on L’sc activity.In contrast to posterior segments, in which ac and sc

are expressed in dorsal and ventral neuroectoderm (Skeath et al.,1994), in the anterior MN expression of both genes alsocovers intermediate neuroectoderm, suggesting a segment-specificregulation by DV genes. It has been shown that Ind is a directrepressor of ac in truncal segments (Zhao et al., 2007a) and that Mshis a repressor of ind expression (Zhao et al., 2007b; Seibert et al.,2009). We found that, specifically in the anterior MN, msh is alsoexpressed early in the intermediate neuroectoderm where it keepsind suppressed and, thereby, enables expression of ac (Fig. 6L,N).Accordingly, in msh68 mutants ind is derepressed (83% ofhemisegments, n=36) and ac is repressed in the anterior-intermediate mandibular neuroectoderm (70% of hemisegments,

n=28) (Fig. 6M,O). Additionally, upon overexpression of ind (usingthe maternal driver Ngt40-Gal4), ac expression was significantlyreduced in the neuroectoderm (82% of hemisegments, n=40;Fig. 6P,Q), often followed by a variable loss of Dpn+ NBs in rows 3,4, 5 and the MP2s (50% of hemisegments, n=40; Fig. 6R,S). Thus,segmental modifications in DV gene expression establish anexpanded domain of ac expression to enable the rapid andconsecutive formation of many NBs from the comparatively smallneuroectoderm in the anterior MN.

Fig. 6. Proneural gene expression in gnathal segments and its segment-specific modification by the activity of DV genes (ind, msh). (A) Stage 8,showing delayed onset of ac expression in MN. (B,C) During stage 9/10 ac isexpressed in a large domain in MN. (D) Expression of sc is similar to that of ac.(E,F) At stage 11 l’sc expression is downregulated in MX and LB, but isprominent in anterior MN. Black lines (A,B,D,E) indicate the segmental borderof MN. (G-L) Stage 10. Black dashed lines indicate segmental borders andwhite dotted frames indicate corresponding regions of intermediateneuroectoderm. Ectopic ind expression in intermediate mandibularneuroectoderm in msh68 (H) or Ngt40>ind (K) embryos represses ac (J,K).(L) Summary of G-K (see main text for details). (M,N) Composite confocalimages of Dpn and En labeling in wild-type (M) and Ngt40>ind (N) embryos atlate stage 11. Upon ind overexpression many row 3, 4, 5 NBs and MP2s(arrows), which would normally develop from the mandibular ac expressiondomain (see Fig. S2.6A′,B′), are lacking.

1297


DEVELO

PM

ENT

http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1

DISCUSSIONNB maps of the gnathal segments complete thecharacterization of segmental patterns and individualidentities for all NBs building the CNS of the flyWhereas substantial knowledge has accumulated regarding thecharacteristics of neural stem cells and the generation of celldiversity in the truncal (thoracic/abdominal) CNS, relatively little isknown about the developing gnathal neuromeres, which represent apeculiar transitional tissue in that they form the most anterior part ofthe larval VNC which becomes associated with the adult brain. Inthis study we describe the spatiotemporal pattern of NB formation inthe gnathal segments (stages 8 to 11) and provide the firstcomprehensive maps of 46 marker genes expressed in subsets ofNBs by late stage 11. At that stage the segmental units of head andtrunk are most clearly displayed, and the final pattern of NBs isestablished. With the NB maps presented here for the gnathalregion, characterization of the entire population of neural stem cellsbuilding the VNC and brain (except the optic lobes) of the fly iscompleted (in total 2×567 NBs), each of these cells now beingindividually identified (Doe, 1992; Urbach et al., 2003; Birkholzet al., 2013a; this study, see Figs 4 and 7); this number, however,does not include the mesectodermal midline progenitors (Bossingand Technau, 1994; Bossing and Brand, 2006; Wheeler et al.,2006), which remain to be determined in the gnathal and pregnathalsegments (except MNB, this study).Remarkably, almost all identified gnathal NBs represent serial

homologs of NBs in the thoracic/abdominal segments, and some ofthem of NBs in the brain (as discussed below), which is reflected bythe neuroectodermal position and time point of their formation, bythe NB-specific code of identity genes (see Table S2) and, as shownfor some of the gnathal NBs, by the production of typical cell lineagecomponents. As shown in previous studies, some seriallyhomologous thoracic/abdominal NBs produce modified segment-specific lineages due to differential specification of NBs and theirprogeny, differences in proliferation and/or PCD (for reviews, seeRogulja-Ortmann and Technau, 2008; Technau et al., 2006). For theNB4-2 lineage, we show that, departing from the situation in the

thorax, the RP2 motoneuron undergoes PCD in the late embryonicLB and MX, whereas it seems not to be formed in MN.Concomitantly, the NB4-2-specific marker gene code revealsdifferences in each gnathal segment in terms of Dfd, Sex combsreduced (Scr) and cap-n-collar (cnc) expression; further,CenG1A istagma-specifically expressed in gnathal NB4-2, and cas and chrbonly inmandibular NB4-2. These divergently expressed factors mayrepresent candidates that confer the segment-specific characteristicsto their lineages. Modifications in the expression profile are foundamong many serially homologous gnathal NBs, but particularly inmandibular NBs. Therefore, we expect lineage divergences to bemost pronounced in MN.

The completed NB map sets the stage for systematic cell lineageanalysis. By comparison with the known lineages of the thoracicand anterior abdominal neuromeres, such an analysis will uncoverthe degree of conservation of serially homologous lineages inderived segments. It will also allow an investigation of the impact ofdifferences in the molecular signature of their stem cells on thepresence or absence of particular progeny cells.

Regulation of the segment-specific patterns of gnathal NBsHox genes play a crucial role in segmental identity and patterning, inpart by regulating PCD (reviewed by Rogulja-Ortmann andTechnau, 2008; Reichert and Bello, 2010). In gnathal segments,PCD has been reported to occur at segmental boundaries (Nassifet al., 1998). It is required for the maintenance of a normal boundarybetween maxillary and mandibular lobes and is induced by Dfd,which activates the pro-apoptosis gene reaper (Lohmann et al.,2002). Here, we have shown that Dfd suppresses the formation ofparticular gnathal NBs in different ways: in case of mandibularNB6-4 by inducing localized PCD in neuroectodermal progenitorsthat constitute the NB6-4 proneural cluster; in case of maxillaryNB5-4 in a PCD-independent manner, possibly by repressing NB-promoting genes (i.e. proneural genes). So far we do not know if Scrplays a role in suppressing NB formation in the labial segment.Recently, the Hox genes Abdominal B and caudal have beenreported to suppress the development of specific NBs in the terminal

Fig. 7. Comparison of the NB pattern in gnathal,thoracic and abdominal segments. The schemeillustrates NBs that are formed (colored) or missing(X) in the left hemineuromere of mandibular (MN),maxillary (MX) or labial (LB) segment and terminalabdominal segments (A8, A9, A10; according toBirkholz et al., 2013a), as comparedwith the groundstate pattern (according to Doe, 1992), which isrepeated in all segments from T1 to A7.

1298


DEVELO

PM

ENT


abdominal segment in a PCD-independent manner (Birkholz et al.,2013b).Only a few NBs are restored in Dfd16 mutants (NB6-4 in MN,

NBs 2-4, 5-4 inMX, NB2-4 in LB). These do not include NBs in thestrongly reduced anterior compartment of MN. Since Dfd protein isdownregulated until stage 11 specifically in the neuroectoderm ofthis compartment (Fig. 5I) (see also McGinnis et al., 1998), it mightnot affect the formation of the comparatively late-developinganterior NBs. As PCD also plays only a minor role, it is likely thatNB numbers are mainly determined by the smaller size of gnathalsegmental anlagen (for blastodermal fate map see Hartenstein et al.,1985), which include lower numbers of neuroectodermalprogenitors, especially in the anterior MN, where the decrease inNBs is most apparent. Different sizes of the gnathalneuroectodermal anlagen are assumed to be defined by earlypatterning genes acting along the DV (e.g. dpp, sog, dorsal) and AP(e.g. gap genes hb, btd, gt) axes (Cohen and Jürgens, 1990; Reinitzand Levine, 1990; Stathopoulos et al., 2002; Mizutani et al., 2005).Finally, we cannot exclude the possibility that homeotic genes (i.e.Dfd, Scr) are also involved, although the size of the early gnathalneuroectoderm seems unaltered in Dfd mutants.Despite the substantial reduction in the number of

neuroectodermal progenitor cells in the anterior MN, many NBsdevelop within a short time window from an oversized proneuraldomain that constantly expresses high levels of ac and sc.Considering its extent and the relatively high number (eight tonine) NBs emerging from this proneural domain, we assume thatthese NBs originate from adjacent neuroectodermal progenitor cells,similar to the situation in the central brain (Urbach et al., 2003; Kunzet al., 2012). This contrasts with the situation inMX, LB and truncalsegments, where proneural clusters are small andwhere only one cellper cluster adopts an NB fate. Thus, spatial and temporal patterns ofproneural gene expression, and presumably the modes of NBformation, differ between gnathal segments. We show that theoversized ac domain in MN is established by a segment-specificalteration in DV gene expression. It is likely that other factors arealso involved in the regionalization and maintenance of ac and scexpression. For example, the cephalic gap gene btd (Younossi-Hartenstein et al., 1997), gt (expressed similarly to ac in anteriorMN; Fig. S2.1D), cnc or col (both specifically expressed in MN;McGinnis et al., 1998; Crozatier et al., 1999) are potential regulators.

Segmental deviation from the NB ground state pattern –comparison of gnathal and terminal abdominal segmentsThe pattern of embryonic NBs in the thoracic (T1-T3) and anteriorabdominal (A1-A7) segments resembles the ground state (T2)(Lewis, 1978). Segmental divergence from this ground pattern isobvious in all gnathal (this study) and terminal abdominal (A8-A10)(Birkholz et al., 2013a) segments, as summarized in Fig. 7. In thegnathal segments the population of NBs is progressively diminishedin the anterior direction, i.e. from LB toMN (∼3,∼5 and 12 NBs aremissing in LB, MX and MN, respectively). In the abdominal tailregion it is progressively diminished in the posterior direction, i.e.from A8 to A10 (one, nine and 21 NBs are missing in A8, A9 andA10, respectively). Notably, in the terminal abdominal segments(A9, A10) NBs are lacking preferentially in the posteriorcompartment (Birkholz et al., 2013a), whereas in the gnathalregion NBs aremissing preferentially in the anterior compartment ofeach segment. Nevertheless, some NBs appear to be preferredvictims of segmental modification in gnathal as well as in terminalabdominal segments: NB2-3 is missing in all gnathal segments andin A8-A10, and NB2-4 is missing in all gnathal segments and A9,

A10. Conversely, seven NBs can be consistently identified in all ofthese segments (NBs 2-1, 2-2, 3-4, 3-5, 5-2, 5-3, 5-6), and thus aremore resistant to segmental modifications in NB patterns.Nevertheless, several of these ‘conserved’ NBs generate lineagesthat have been shown to differ among gnathal, thoracic andabdominal segments with regard to the number and/or particulartypes of progeny cells, and there are indications that thesedifferences primarily affect the later parts of their lineages(Bossing et al., 1996; Schmidt et al., 1997; Schmid et al., 1999;Baumgardt et al., 2009; Karlsson et al., 2010). Thus, the early partsof their lineages might form repetitive units of the neuronal network,the specific functions of which might be required in all truncal andgnathal neuromeres.

Linking postembryonic lineages to embryonic NBs in thegnathal CNSAfter a period of mitotic quiescence, a specific subset of embryonicNBs resumes proliferation during larval development to produceadult-specific neurons (reviewed by Maurange and Gould, 2005). Inthoracic and abdominal segments, embryonic NBs are reactivated in asegment-specific pattern (Truman and Bate, 1988), which is alreadydetermined by Hox genes in the embryonic neuroectoderm (Prokopet al., 1998). In the gnathal segments, a substantial number (>80%) ofembryonic NBs disappears during late embryonic and larvaldevelopment, most of them due to PCD. Only ∼14 NBs and theircorresponding lineages (13 paired, 1 unpaired) have been detected inthe gnathal region of the late larva, most of which develop in LB(Kuert et al., 2014). Each of the eight to nine postembryonic lineagesin LB could already be linked to a specific embryonic NB, whichdemonstrated that in LB and abdominal segment A1 (with a likewisereduced set of reactivated NBs) they emerge from a similar set ofserially homologous embryonic NBs (Birkholz et al., 2015). Serialhomologs of most of these NBs also exist in embryonic MX andMN(e.g. NBs 4-2, 5-2, 6-1, 6-2, 7-1, 7-4, MNB), which are likely toinclude the precursors of the remaining (approximately six)postembryonic lineages described by Kuert et al. (2014) for thegnathal region. As four of those lineages express En (Kuert et al.,2014), they presumably stem from embryonic En-expressing NBs inrows 6 and 7 (NBs 6-1, 6-2, 7-1, 7-4).

The postembryonic lineages in the VNC of the larva can beindividually identified based on their morphological characteristics(Truman et al., 2004, 2010) and specific codes of gene expression(e.g. Lacin et al., 2014; Li et al., 2014), and all of them have beenlinked to identified embryonic NBs (Birkholz et al., 2015). Notably,sets of transcription factors found to be expressed in specificpostembryonic lineages (Lacin et al., 2014) often differ from thoseexpressed in their parent embryonic NBs, revealing highly dynamicexpression of these factors.

Serially homologous NBs in neuromeres of the abdomen/thorax, gnathal and pregnathal headIn a previous report we gave a first example for a serially homologousNB in neuromeres of trunk (abdomen/thorax) and brain (Urbachet al., 2003). Here we elaborated on these investigations bycomparing the characteristics of all NBs in neuromeres of thethorax, gnathal and pregnathal head, the latter giving rise to the brain(trito-, deuto- and protocerebrum). Remarkably, almost all NBs inthe tritocerebrum seem to represent serial homologs of NBs in thegnathal and thoracic/abdominal segments. Furthermore, thepopulation of tritocerebral NBs is likely to represent a subset ofNBs found inMN and, in turn, those in MN,MX and LB represent asubset of those in the next posterior segment (MX, LB and T1),

1299


DEVELO

PM

ENT


respectively. This suggests that intersegmental modification of theNB pattern in the three gnathal and the tritocerebral neuromeresinvolves a progressive reduction compared with the thoracic groundstate. Also, approximately half of the deutocerebral NBs displaypotential serial homology to NBs in the gnathal and thoracicneuromeres, but not all of them have a counterpart in thetritocerebrum. Notably, serial homology is not overt in thepopulation of (∼70) protocerebral NBs that constitute the anteriorand largest part of the brain. This is in agreement with recent datashowing that expression patterns of postembryonic lineages in theVNC are not consistently linked to the expression of lineages in thecentral brain (Li et al., 2014). Together, the appearance of seriallyhomologous NBs seems to mirror the order in which segmentalpatterning of the neuroectoderm, and the formation and specificationof NBs, is genetically conserved: conformity to the thoracic groundstate is highest in LB and progressively diverges from MX to theprotocerebrum (this study; Crozatier et al., 1999; Seibert et al., 2009;reviewed by Urbach and Technau, 2004, 2008). The existence ofNBs in the brain showing serial homology to those in the VNC isastounding considering that NB formation is significantly delayed inMN and tritocerebrum (Fig. 1) (Urbach et al., 2003), and thatsegmental patterning of the neuroectoderm as well as the formationof NBs is regulated by different mechanisms in the brain (Hirth et al.,1995; Younossi-Hartenstein et al., 1997; Seibert et al., 2009; Urbachand Technau, 2004).These intersegmental comparisons not only uncover serial

homologies among NBs based on similarities, but also segmentalmodifications in their expression profiles. Since nearly all of themarker genes considered in this study encode transcription factors,these modifications are likely to be responsible for the establishmentof segment-specific characteristics of serially homologous NBlineages. Thus, the NB map forms a basis for comprehensiveanalyses of the lineages building the CNS of the fly and presentscandidate factors controlling their region-specific composition.Some of these factors are presumably part of the genetic networkthat provides instructions for the establishment of neural circuitsinvolved in the processing of feeding and taste response. This workmight therefore also help to decipher the genetic mechanisms thatdirect gnathal-specific behavior.For the first time the entire complement ofNBs in acomplex animal

has been mapped. Combined with the detailed gene expressionstudies in this and previous reports this provides a framework forunderstanding the mechanisms underlying segmental patterning oftheCNS. Itwill also serve as a reference and provide valuable tools forcomparative analyses of neurogenesis in other species.

MATERIALS AND METHODSFly strainsThe fly strains used in this study are listed in the supplementary Materialsand Methods.

Immunohistochemistry and in situ hybridizationEmbryos were dechorionated, fixed, immunostained and flat preparationsperformed according to published protocols (Patel, 1994; Jussen andUrbach, 2014). The antibodies used are detailed in the supplementaryMaterials and Methods. Probes for in situ hybridization were synthesizedusing the templates shown in Table S4. In situ hybridization was performedas described previously (Jussen and Urbach, 2014) and probes processedwith NBT/BCIP solution for non-fluorescent staining (Carl Roth) ortyramide signal amplification (TSA Cyanine 3 System; PerkinElmer) forfluorescent stainings. Embryos were then immunolabeled with primaryantibody followed by incubation with biotinylated (processed with DAB)or fluorescent dye-coupled secondary antibodies as described in the

supplementary Materials and Methods. Non-fluorescent stainings weredocumented on a Zeiss Axioplan microscope, while fluorescent confocalimages were acquired on a Leica TCS SP5 II microscope. Images wereprocessed with ImageJ (NIH), Adobe Photoshop and Adobe Illustrator.

AcknowledgementsWe thank Dagmar Volland for technical assistance and are very grateful to RalfPflanz, Uwe Walldorf, Ward Odenwald, Tonia von Ohlen, Joachim Urban, JamesSkeath, Chris Doe, John Reinitz, Tiffany Cook, Christian Klämbt, Andrew Jarman,Miche ̀le Crozatier, Jürgen Knoblich, Manfred Frasch, Robert Holmgren, Ethan Bier,Krzysztof Jagla, Laura Nilson, Matthew Scott, Ze’ev Paroush, Takako Isshiki,Markus Noll, Fernando Jimenez, Harald Vaessin, Yu Cai, Benjamin Altenhein,Matthias Landgraf, the Bloomington Stock Center, and the Developmental StudiesHybridoma Bank (University of Iowa) for providing flies, antibodies and cDNA.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsR.U. and G.M.T. conceived the study; R.U. designed experiments with input fromD.J. and G.M.T.; D.J. performed the majority of experiments; R.U. analyzed themajority of data; R.U. and D.J. prepared the figures; R.U. and G.M.T. wrote themanuscript.

FundingThis work was supported by grants from the Deutsche Forschungsgemeinschaft[UR163/2-2, UR163/3-3 to R.U. and TE130/10-2, TE130/9-3 to G.M.T.]. Depositedin PMC for immediate release.

Supplementary informationSupplementary information available online athttp://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1

ReferencesBaumgardt, M., Karlsson, D., Terriente, J., Dıáz-Benjumea, F. J. and Thor, S.

(2009). Neuronal subtype specification within a lineage by opposing temporalfeed-forward loops. Cell 139, 969-982.

Birkholz, O., Rickert, C., Berger, C., Urbach, R. and Technau, G. M. (2013a).Neuroblast pattern and identity in the Drosophila tail region and role of doublesexin the survival of sex-specific precursors. Development 140, 1830-1842.

Birkholz, O., Vef, O., Rogulja-Ortmann, A., Berger, C. and Technau, G. M.(2013b). Abdominal-B and caudal inhibit the formation of specific neuroblasts inthe Drosophila tail region. Development 140, 3552-3564.

Birkholz,O.,Rickert,C.,Nowak, J.,Coban, I.C.andTechnau,G.M. (2015).Bridgingthegapbetweenpostembryonic cell lineagesand identifiedembryonicneuroblasts inthe ventral nerve cord of Drosophila melanogaster. Biol. Open 4, 420-434.

Bossing, T. and Brand, A. H. (2006). Determination of cell fate along theanteroposterior axis of the Drosophila ventral midline. Development 133,1001-1012.

Bossing, T. and Technau, G. M. (1994). The fate of the CNS midline progenitors inDrosophila as revealed by a new method for single cell labelling. Development120, 1895-1906.

Bossing, T., Udolph, G., Doe, C. Q. and Technau, G. M. (1996). The embryoniccentral nervous system lineages of Drosophila melanogaster. I. Neuroblastlineages derived from the ventral half of the neuroectoderm.Dev. Biol. 179, 41-64.

Broadus, J., Skeath, J. B., Spana, E. P., Bossing, T., Technau, G. and Doe, C. Q.(1995). New neuroblast markers and the origin of the aCC/pCC neurons in theDrosophila central nervous system. Mech. Dev. 53, 393-402.

Cohen, S. M. and Jürgens, G. (1990). Mediation of Drosophila head developmentby gap-like segmentation genes. Nature 346, 482-485.

Crozatier, M., Valle, D., Dubois, L., Ibnsouda, S. and Vincent, A. (1999). Headversus trunk patterning in the Drosophila embryo; collier requirement for formationof the intercalary segment. Development 126, 4385-4394.

DiNardo, S., Kuner, J. M., Theis, J. and O’Farrell, P. H. (1985). Development ofembryonic pattern in D. melanogaster as revealed by accumulation of the nuclearengrailed protein. Cell 43, 59-69.

Doe, C. Q. (1992). Molecular markers for identified neuroblasts and ganglionmothercells in the Drosophila central nervous system. Development 116, 855-863.

Gendre, N., Lüer, K., Friche, S., Grillenzoni, N., Ramaekers, A., Technau, G. M.and Stocker, R. F. (2004). Integration of complex larval chemosensory organsinto the adult nervous system of Drosophila. Development 131, 83-92.

Goulding, S. E.,White, N. M. and Jarman, A. P. (2000). cato encodes a basic helix-loop-helix transcription factor implicated in the correct differentiation of Drosophilasense organs. Dev. Biol. 221, 120-131.

Hartenstein, V. and Campos-Ortega, J. A. (1984). Early neurogenesis in wild-typeDrosophila melanogaster. Roux’s Arch. Dev. Biol. 193, 308-325.

1300


DEVELO

PM

ENT

http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.133546/-/DC1http://dx.doi.org/10.1016/j.cell.2009.10.032http://dx.doi.org/10.1016/j.cell.2009.10.032http://dx.doi.org/10.1016/j.cell.2009.10.032http://dx.doi.org/10.1242/dev.090043http://dx.doi.org/10.1242/dev.090043http://dx.doi.org/10.1242/dev.090043http://dx.doi.org/10.1242/dev.096099http://dx.doi.org/10.1242/dev.096099http://dx.doi.org/10.1242/dev.096099http://dx.doi.org/10.1242/bio.201411072http://dx.doi.org/10.1242/bio.201411072http://dx.doi.org/10.1242/bio.201411072http://dx.doi.org/10.1242/dev.02288http://dx.doi.org/10.1242/dev.02288http://dx.doi.org/10.1242/dev.02288http://dx.doi.org/10.1006/dbio.1996.0240http://dx.doi.org/10.1006/dbio.1996.0240http://dx.doi.org/10.1006/dbio.1996.0240http://dx.doi.org/10.1016/0925-4773(95)00454-8http://dx.doi.org/10.1016/0925-4773(95)00454-8http://dx.doi.org/10.1016/0925-4773(95)00454-8http://dx.doi.org/10.1038/346482a0http://dx.doi.org/10.1038/346482a0http://dx.doi.org/10.1016/0092-8674(85)90012-1http://dx.doi.org/10.1016/0092-8674(85)90012-1http://dx.doi.org/10.1016/0092-8674(85)90012-1http://dx.doi.org/10.1242/dev.00879http://dx.doi.org/10.1242/dev.00879http://dx.doi.org/10.1242/dev.00879http://dx.doi.org/10.1006/dbio.2000.9677http://dx.doi.org/10.1006/dbio.2000.9677http://dx.doi.org/10.1006/dbio.2000.9677http://dx.doi.org/10.1007/BF00848159http://dx.doi.org/10.1007/BF00848159

Hartenstein, V., Technau, G. M. and Campos-Ortega, J. A. (1985). Fate-mappingin wild-type Drosophila melanogaster. III. Fate map of the blastoderm. Roux’sArch. Dev. Biol. 194, 213-216.

Hartenstein, V., Rudolff, E. and Campos-Ortega, J. A. (1987). The pattern ofproliferation of the neuroblasts in the wild-type embryo of Drosophilamelanogaster. Roux’s Arch. Dev. Biol. 196, 473-485.

Hirth, F., Therianos, S., Loop, T., Gehring, W. J., Reichert, H. and Furukubo-Tokunaga, K. (1995). Developmental defects in brain segmentation caused bymutations of the homeobox genes orthodenticle and empty spiracles inDrosophila. Neuron 15, 769-778.

Ito, K., Shinomiya, K., Ito, M., Armstrong, J. D., Boyan, G., Hartenstein, V.,Harzsch, S., Heisenberg, M., Homberg, U., Jenett, A. et al. (2014). Asystematic nomenclature for the insect brain. Neuron 81, 755-765.

Jiménez, F. and Campos-Ortega, J. A. (1990). Defective neuroblast commitmentin mutants of the achaete-scute complex and adjacent genes of Drosophilamelanogaster. Neuron 5, 81-89.

Jussen, D. and Urbach, R. (2014). Non-fluorescent RNA in situ hybridizationcombined with antibody staining to visualize multiple gene expression patterns inthe embryonic brain of Drosophila. Methods Mol. Biol. 1082, 19-35.

Karlsson, D., Baumgardt, M. and Thor, S. (2010). Segment-specific neuronalsubtype specification by the integration of anteroposterior and temporal cues.PLoS Biol. 8, e1000368.

Kuert, P. A., Hartenstein, V., Bello, B. C., Lovick, J. K. and Reichert, H. (2014).Neuroblast lineage identification and lineage-specific Hox gene action duringpostembryonic development of the subesophageal ganglion in the Drosophilacentral brain. Dev. Biol. 390, 102-115.

Kunz, T., Kraft, K. F., Technau, G. M. and Urbach, R. (2012). Origin of Drosophilamushroom body neuroblasts and generation of divergent embryonic lineages.Development 139, 2510-2522.

Lacin, H., Zhu, Y., Wilson, B. A. and Skeath, J. B. (2014). Transcription factorexpression uniquely identifies most postembryonic neuronal lineages in theDrosophila thoracic central nervous system. Development 141, 1011-1021.

Lewis, E. B. (1978). A gene complex controlling segmentation in Drosophila.Nature276, 565-570.

Li, H.-H., Kroll, J. R., Lennox, S. M., Ogundeyi, O., Jeter, J., Depasquale, G. andTruman, J. W. (2014). A GAL4 driver resource for developmental and behavioralstudies on the larval CNS of Drosophila. Cell Rep. 8, 897-908.

Lohmann, I., McGinnis, N., Bodmer, M. andMcGinnis, W. (2002). The DrosophilaHox gene deformed sculpts head morphology via direct regulation of theapoptosis activator reaper. Cell 110, 457-466.

Maurange, C. and Gould, A. P. (2005). Brainy but not too brainy: starting andstopping neuroblast divisions in Drosophila. Trends Neurosci. 28, 30-36.

McGinnis, N., Ragnhildstveit, E., Veraksa, A. and McGinnis, W. (1998). A cap ‘n’collar protein isoform contains a selective Hox repressor function. Development125, 4553-4564.

Melcher, C. and Pankratz, M. J. (2005). Candidate gustatory interneuronsmodulating feeding behavior in the Drosophila brain. PLoS Biol. 3, e305.

Mizutani, C. M., Nie, Q., Wan, F. Y. M., Zhang, Y.-T., Vilmos, P., Sousa-Neves, R.,Bier, E., Marsh, J. L. and Lander, A. D. (2005). Formation of the BMP activitygradient in the Drosophila embryo. Dev. Cell 8, 915-924.

Nassif, C., Daniel, A., Lengyel, J. A. and Hartenstein, V. (1998). The role ofmorphogenetic cell death during Drosophila embryonic head development. Dev.Biol. 197, 170-186.

Patel, N. H. (1994). Imaging neuronal subsets and other cell types in whole-mountDrosophila embryos and larvae using antibody probes. Methods Cell Biol. 44,445-487.

Prokop, A., Bray, S., Harrison, E. and Technau, G. M. (1998). Homeotic regulationof segment-specific differences in neuroblast numbers and proliferation in theDrosophila central nervous system. Mech. Dev. 74, 99-110.

Reichert, H. and Bello, B. (2010). Hox genes and brain development in Drosophila.Adv. Exp. Med. Biol. 689, 145-153.

Reinitz, J. and Levine, M. (1990). Control of the initiation of homeotic geneexpression by the gap genes giant and tailless in Drosophila. Dev. Biol. 140,57-72.

Rogulja-Ortmann, A. and Technau, G. M. (2008). Multiple roles for Hox genes insegment-specific shaping of CNS lineages. Fly 2, 316-319.

Schmid, A., Chiba, A. and Doe, C. Q. (1999). Clonal analysis of Drosophilaembryonic neuroblasts: neural cell types, axon projections and muscle targets.Development 126, 4653-4689.

Schmidt, H., Rickert, C., Bossing, T., Vef, O., Urban, J. and Technau, G. M.(1997). The embryonic central nervous system lineages of Drosophilamelanogaster. II. Neuroblast lineages derived from the dorsal part of theneuroectoderm. Dev. Biol. 189, 186-204.

Schmidt-Ott, U. and Technau, G. M. (1992). Expression of en and wg in theembryonic head and brain of Drosophila indicates a refolded band of sevensegment remnants. Development 116, 111-125.

Scott, K., Brady, R., Jr, Cravchik, A., Morozov, P., Rzhetsky, A., Zuker, C. andAxel, R. (2001). A chemosensory gene family encoding candidate gustatory andolfactory receptors in Drosophila. Cell 104, 661-673.

Seibert, J., Volland, D. and Urbach, R. (2009). Ems and Nkx6 are centralregulators in dorsoventral patterning of the Drosophila brain. Development 136,3937-3947.

Skeath, J. B. and Thor, S. (2003). Genetic control of Drosophila nerve corddevelopment. Curr. Opin. Neurobiol. 13, 8-15.

Skeath, J. B., Panganiban, G. F. and Sean, B. and Carroll, S. B. (1994). Theventral nervous system defective gene controls proneural gene expression at twodistinct steps during neuroblast formation in Drosophila. Development 120,1517-1524.

Spana, E. P. and Doe, C. Q. (1995). The prospero transcription factor isasymmetrically localized to the cell cortex during neuroblast mitosis inDrosophila. Development 121, 3187-3195.

Stathopoulos, A., Van Drenth, M., Erives, A., Markstein, M. and Levine, M.(2002). Whole-genome analysis of dorsal-ventral patterning in the Drosophilaembryo. Cell 111, 687-701.

Technau, G. M., Berger, C. and Urbach, R. (2006). Generation of cell diversity andsegmental pattern in the embryonic central nervous system of Drosophila. Dev.Dyn. 235, 861-869.

Truman, J. W. and Bate, M. (1988). Spatial and temporal patterns of neurogenesisin the central nervous system of Drosophila melanogaster. Dev. Biol. 125,145-157.

Truman, J. W., Schuppe, H., Shepherd, D. and Williams, D. W. (2004).Developmental architecture of adult-specific lineages in the ventral CNS ofDrosophila. Development 131, 5167-5184.

Truman, J. W., Moats, W., Altman, J., Marin, E. C. and Williams, D. W. (2010).Role of Notch signaling in establishing the hemilineages of secondary neurons inDrosophila melanogaster. Development 137, 53-61.

Urbach, R. (2007). A procephalic territory in Drosophila exhibiting similarities anddissimilarities compared to the vertebrate midbrain/hindbrain boundary region.Neural Dev. 2, 23.

Urbach, R. and Technau, G. M. (2003a). Segment polarity and DV patterning geneexpression reveals segmental organization of the Drosophila brain. Development130, 3607-3620.

Urbach, R. and Technau, G. M. (2003b). Molecular markers for identifiedneuroblasts in the developing brain of Drosophila. Development 130, 3621-3637.

Urbach, R. and Technau, G. M. (2004). Neuroblast formation and patterning duringearly brain development in Drosophila. Bioessays 26, 739-751.

Urbach, R. and Technau, G. M. (2008). Dorsoventral patterning of the brain – acomparative approach. In Advances in Experimental Medicine and Biology, Vol.628 (ed. G. M. Technau), pp. 42-56. Austin: Landes Bioscience.

Urbach, R., Schnabel, R. and Technau, G. M. (2003). The pattern of neuroblastformation, mitotic domains and proneural gene expression during early braindevelopment in Drosophila. Development 130, 3589-3606.

Weiss, J. B., Von Ohlen, T., Mellerick, D. M., Dressler, G., Doe, C. Q. and Scott,M. P. (1998). Dorsoventral patterning in the Drosophila central nervous system:the intermediate neuroblasts defective homeobox gene specifies intermediatecolumn identity. Genes Dev. 12, 3591-3602.

Wheeler, S. R., Kearney, J. B., Guardiola, A. R. and Crews, S. T. (2006). Single-cell mapping of neural and glial gene expression in the developing DrosophilaCNS midline cells. Dev. Biol. 294, 509-524.

White, K., Grether, M. E., Abrams, J. M., Young, L., Farrell, K. and Steller, H.(1994). Genetic control of programmed cell death in Drosophila. Science 264,677-683.

Younossi-Hartenstein, A., Nassif, C., Green, P. and Hartenstein, V. (1996). Earlyneurogenesis of the Drosophila brain. J. Comp. Neurol. 370, 313-329.

Younossi-Hartenstein, A., Green, P., Liaw, G.-J., Rudolph, K., Lengyel, J. andHartenstein, V. (1997). Control of early neurogenesis of the Drosophila brain bythe head gap genes tll, otd, ems, and btd. Dev. Biol. 182, 270-283.

Zhao, G., Boekhoff-Falk, G., Wilson, B. A. and Skeath, J. B. (2007a). Linkingpattern formation to cell-type specification: Dichaete and Ind directly repressachaete gene expression in the Drosophila CNS. Proc. Natl. Acad. Sci. USA 104,3847-3852.

Zhao, G., Wheeler, S. R. and Skeath, J. B. (2007b). Genetic control of dorsoventralpatterning and neuroblast specification in the Drosophila central nervous system.Int. J. Dev. Biol. 51, 107-115.

1301


DEVELO

PM

ENT

http://dx.doi.org/10.1007/BF00848248http://dx.doi.org/10.1007/BF00848248http://dx.doi.org/10.1007/BF00848248http://dx.doi.org/10.1007/BF00399871http://dx.doi.org/10.1007/BF00399871http://dx.doi.org/10.1007/BF00399871http://dx.doi.org/10.1016/0896-6273(95)90169-8http://dx.doi.org/10.1016/0896-6273(95)90169-8http://dx.doi.org/10.1016/0896-6273(95)90169-8http://dx.doi.org/10.1016/0896-6273(95)90169-8http://dx.doi.org/10.1016/j.neuron.2013.12.017http://dx.doi.org/10.1016/j.neuron.2013.12.017http://dx.doi.org/10.1016/j.neuron.2013.12.017http://dx.doi.org/10.1016/0896-6273(90)90036-Fhttp://dx.doi.org/10.1016/0896-6273(90)90036-Fhttp://dx.doi.org/10.1016/0896-6273(90)90036-Fhttp://dx.doi.org/10.1007/978-1-62703-655-9_2http://dx.doi.org/10.1007/978-1-62703-655-9_2http://dx.doi.org/10.1007/978-1-62703-655-9_2http://dx.doi.org/10.1371/journal.pbio.1000368http://dx.doi.org/10.1371/journal.pbio.1000368http://dx.doi.org/10.1371/journal.pbio.1000368http://dx.doi.org/10.1016/j.ydbio.2014.03.021http://dx.doi.org/10.1016/j.ydbio.2014.03.021http://dx.doi.org/10.1016/j.ydbio.2014.03.021http://dx.doi.org/10.1016/j.ydbio.2014.03.021http://dx.doi.org/10.1242/dev.077883http://dx.doi.org/10.1242/dev.077883http://dx.doi.org/10.1242/dev.077883http://dx.doi.org/10.1242/dev.102178http://dx.doi.org/10.1242/dev.102178http://dx.doi.org/10.1242/dev.102178http://dx.doi.org/10.1038/276565a0http://dx.doi.org/10.1038/276565a0http://dx.doi.org/10.1016/j.celrep.2014.06.065http://dx.doi.org/10.1016/j.celrep.2014.06.065http://dx.doi.org/10.1016/j.celrep.2014.06.065http://dx.doi.org/10.1016/S0092-8674(02)00871-1http://dx.doi.org/10.1016/S0092-8674(02)00871-1http://dx.doi.org/10.1016/S0092-8674(02)00871-1http://dx.doi.org/10.1016/j.tins.2004.10.009http://dx.doi.org/10.1016/j.tins.2004.10.009http://dx.doi.org/10.1371/journal.pbio.0030305http://dx.doi.org/10.1371/journal.pbio.0030305http://dx.doi.org/10.1016/j.devcel.2005.04.009http://dx.doi.org/10.1016/j.devcel.2005.04.009http://dx.doi.org/10.1016/j.devcel.2005.04.009http://dx.doi.org/10.1006/dbio.1998.8875http://dx.doi.org/10.1006/dbio.1998.8875http://dx.doi.org/10.1006/dbio.1998.8875http://dx.doi.org/10.1016/S0091-679X(08)60927-9http://dx.doi.org/10.1016/S0091-679X(08)60927-9http://dx.doi.org/10.1016/S0091-679X(08)60927-9http://dx.doi.org/10.1016/S0925-4773(98)00068-9http://dx.doi.org/10.1016/S0925-4773(98)00068-9http://dx.doi.org/10.1016/S0925-4773(98)00068-9http://dx.doi.org/10.1007/978-1-4419-6673-5_11http://dx.doi.org/10.1007/978-1-4419-6673-5_11http://dx.doi.org/10.1016/0012-1606(90)90053-Lhttp://dx.doi.org/10.1016/0012-1606(90)90053-Lhttp://dx.doi.org/10.1016/0012-1606(90)90053-Lhttp://dx.doi.org/10.4161/fly.7464http://dx.doi.org/10.4161/fly.7464http://dx.doi.org/10.1006/dbio.1997.8660http://dx.doi.org/10.1006/dbio.1997.8660http://dx.doi.org/10.1006/dbio.1997.8660http://dx.doi.org/10.1006/dbio.1997.8660http://dx.doi.org/10.1016/S0092-8674(01)00263-Xhttp://dx.doi.org/10.1016/S0092-8674(01)00263-Xhttp://dx.doi.org/10.1016/S0092-8674(01)00263-Xhttp://dx.doi.org/10.1242/dev.041921http://dx.doi.org/10.1242/dev.041921http://dx.doi.org/10.1242/dev.041921http://dx.doi.org/10.1016/S0959-4388(03)00007-2http://dx.doi.org/10.1016/S0959-4388(03)00007-2http://dx.doi.org/10.1016/S0092-8674(02)01087-5http://dx.doi.org/10.1016/S0092-8674(02)01087-5http://dx.doi.org/10.1016/S0092-8674(02)01087-5http://dx.doi.org/10.1002/dvdy.20566http://dx.doi.org/10.1002/dvdy.20566http://dx.doi.org/10.1002/dvdy.20566http://dx.doi.org/10.1016/0012-1606(88)90067-Xhttp://dx.doi.org/10.1016/0012-1606(88)90067-Xhttp://dx.doi.org/10.1016/0012-1606(88)90067-Xhttp://dx.doi.org/10.1242/dev.01371http://dx.doi.org/10.1242/dev.01371http://dx.doi.org/10.1242/dev.01371http://dx.doi.org/10.1242/dev.041749http://dx.doi.org/10.1242/dev.041749http://dx.doi.org/10.1242/dev.041749http://dx.doi.org/10.1186/1749-8104-2-23http://dx.doi.org/10.1186/1749-8104-2-23http://dx.doi.org/10.1186/1749-8104-2-23http://dx.doi.org/10.1242/dev.00532http://dx.doi.org/10.1242/dev.00532http://dx.doi.org/10.1242/dev.00532http://dx.doi.org/10.1242/dev.00533http://dx.doi.org/10.1242/dev.00533http://dx.doi.org/10.1002/bies.20062http://dx.doi.org/10.1002/bies.20062http://dx.doi.org/10.1242/dev.00528http://dx.doi.org/10.1242/dev.00528http://dx.doi.org/10.1242/dev.00528http://dx.doi.org/10.1101/gad.12.22.3591http://dx.doi.org/10.1101/gad.12.22.3591http://dx.doi.org/10.1101/gad.12.22.3591http://dx.doi.org/10.1101/gad.12.22.3591http://dx.doi.org/10.1016/j.ydbio.2006.03.016http://dx.doi.org/10.1016/j.ydbio.2006.03.016http://dx.doi.org/10.1016/j.ydbio.2006.03.016http://dx.doi.org/10.1126/science.8171319http://dx.doi.org/10.1126/science.8171319http://dx.doi.org/10.1126/science.8171319http://dx.doi.org/10.1002/(SICI)1096-9861(19960701)370:33.0.CO;2-7http://dx.doi.org/10.1002/(SICI)1096-9861(19960701)370:33.0.CO;2-7http://dx.doi.org/10.1006/dbio.1996.8475http://dx.doi.org/10.1006/dbio.1996.8475http://dx.doi.org/10.1006/dbio.1996.8475http://dx.doi.org/10.1073/pnas.0611700104http://dx.doi.org/10.1073/pnas.0611700104http://dx.doi.org/10.1073/pnas.0611700104http://dx.doi.org/10.1073/pnas.0611700104http://dx.doi.org/10.1387/ijdb.062188gzhttp://dx.doi.org/10.1387/ijdb.062188gzhttp://dx.doi.org/10.1387/ijdb.062188gz

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 150 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 200 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.32000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 400 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.00000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError false /PDFXTrimBoxToMediaBoxOffset [ 34.69606 34.27087 34.69606 34.27087 ] /PDFXSetBleedBoxToMediaBox false /PDFXBleedBoxToTrimBoxOffset [ 8.50394 8.50394 8.50394 8.50394 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False

/CreateJDFFile false /Description > /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ > /FormElements false /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles false /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /DocumentCMYK /PreserveEditing true /UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> ]>> setdistillerparams> setpagedevice

gene expression profiles uncover individual identities of …...which lie in close vicinity to...

Documents