mummy encodes an UDP-N-acetylglucosamine-dipohosphorylase and is

required during Drosophila dorsal closure and nervous system development

Kristina Schimmelpfeng 1, Mareike Strunk, Tobias Stork, Christian Klambt *

Institut fur Neurobiologie, Universitat Munster, Badestr. 9, 48149 Munster, Germany

Received 16 December 2005; received in revised form 1 March 2006; accepted 9 March 2006

Available online 27 March 2006

Abstract

Throughout development cell–cell interactions are of pivotal importance. Cells bind to each other or share information via secreted signaling

molecules. To a large degree, these processes are modulated by post-translational modifications of membrane proteins. Glycan-chains are

frequently added to membrane proteins and assist their exact function at the cell surface. In addition, the glycosylation pathway is required to

generate GPI-linkage in the endoplasmatic reticulum. Here, we describe the analysis of the cabrio/mummy gene, which encodes an UDP-N-

acetylglucosamine diphosphorylase. This is a well-conserved and central enzyme in the glycosylation pathway. As expected from this central role

in glycosylation, cabrio/mummy mutants show many phenotypic traits ranging from CNS fasciculation defects to defects in dorsal closure and eye

development. These phenotypes correlate well with specific glycosylation and GPI-anchorage defects in mummy mutants.

q 2006 Elsevier Ireland Ltd. All rights reserved.

Keywords: Drosophila; Axon patterning; Dorsal closure; UDP-N-acetylglucosamine-diphosphorylase

1. Introduction

Cell–cell communication is an important feature common

to all multi-cellular organisms. On the one hand, cells can

communicate via a cohort of extracellular signaling

molecules, which are then perceived by specific receptors.

The extracellular distribution as well as the binding of these

signals to their receptors is regulated in part via

glycosylation (Hacker et al., 2005; Haltiwanger and Stanley,

2002; Lee and Chien, 2004; Lin, 2004). In the developing

wing of Drosophila, the sugar moiety of the heparan sulfate

proteoglycans (HPSG) facilitates the spreading of morpho-

gens such as Decapentaplegic (Dpp), Wingless (Wg) or

Hedgehog (Hh) and thus participate in the generation of

morphogen gradients (Belenkaya et al., 2004; Bornemann

et al., 2004; Han et al., 2004; Takei et al., 2004). Similarly,

HSPGs have been shown to be important Drosophila eye

development (Kirkpatrick et al., 2004; Nakato et al., 2002).

0925-4773/$ - see front matter q 2006 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.mod.2006.03.004

* Corresponding author. Tel.: C49 251 83 2 1122; fax: C49 251 83 2 4686.

E-mail address: klaembt@uni-muenster.de (C. Klambt).1 Present address: Department of Cellular and Molecular Medicine, LBR449,

University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-

0683, USA.

The role of sugar modifications is also evident in the

developing nervous systems where neurons send out axonal

processes to form a highly ordered lattice of intercellular

connections (Holt and Dickson, 2005; Lee and Chien, 2004).

During the formation of this network, the growing tip of the

axon, the growth cone, interprets several cellular signals to

successfully navigate through the developing embryo

(Dickson, 2002). The majority of the neurons in the central

nervous system (CNS) are interneurons that send their axons to

the contra-lateral side of the body. This journey across the CNS

midline is mostly guided by two conserved signaling systems

represented by the ligand/receptor complexes: Netrin/Dcc and

Slit/Robo (Barallobre et al., 2005; Wong et al., 2002). Initially,

the CNS midline cells secrete Netrin, which serves as an

attractive signal steering growth cones towards the midline by

activation of the Deleted in colorectal cancer (Dcc) receptor

protein. Once at the CNS midline the growth cones encounter

the repulsive molecule Slit, which mediates repulsive growth

via activation of the Roundabout receptor family, and thus

ensures that growth cones cross the midline only once.

Following traversing the midline, axons grow in specific

fascicles in the longitudinal axon bundles. The molecular

mechanisms underlying this fasciculation are not known to

date, however, sugar codes on specific receptor proteins may

Mechanisms of Development 123 (2006) 487–499

www.elsevier.com/locate/modo

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499488

serve as a mechanism to ensure the high fidelity of axonal

navigation (Holt and Dickson, 2005; Lee and Chien, 2004).

Indeed, carbohydrate epitopes are found on many neuronally

expressed extracellular proteins (Desai et al., 1994; Snow et al.,

1987; Sun and Salvaterra, 1995).

In several cases, it was shown that carbohydrate moieties are

able to regulate binding activity of receptors. For example the

removal of polysialic acid from neuronal NCAM inhibits

neurite outgrowth of chicken retinal ganglion cells in vitro

(Doherty et al., 1990). Carbohydrate modifications can also

modulate receptor function as seen for Notch, whose activity is

controlled by the O-fucosyltransferase (O-FucT-1) Neurotic

and the UDP-glycosyltransferase Fringe (Bruckner et al., 2000;

Haines and Irvine, 2003; Moloney et al., 2000; Okajima et al.,

2003; Sasamura et al., 2003).

Heparan sulfate proteoglycans have also been implicated in

the regulation of Slit/Round about signaling system in

Drosophila, C. elegans and zebrafish (Bulow et al., 2002;

Bulow and Hobert, 2004; Johnson et al., 2004; Lee et al., 2004;

Nybakken and Perrimon, 2002; Steigemann et al., 2004).

Furthermore, it has been shown, that Drosophila mutations that

interfere with the expression of the HRP-epitope, an N-linked

oligosaccharide expressed on a subset of neuronal glyco-

proteins, lead to defects in neuronal development. Specifically,

mutations in the neurally altered carbohydrate (nac) gene lead

to axon defasciculation and misrouting of sensory neurons

from the wing into the CNS (Whitlock, 1993).

Besides representing extracellular signaling codes, glycosyla-

tion does occur in many other instances. Glycosylation is needed

to generate the GPI anchors that keep some proteins at the outer

leaflet of the plasma membrane (Herscovics and Orlean, 1993). In

addition, the exoskeleton of invertebrates is made of chitin

consisting of poly-N-acetylglucosamine, which is one of the most

abundant biopolymers known (Merzendorfer and Zimoch, 2003).

To identify the genes that are required for normal formation

of the embryonic nervous system of Drosophila we have

previously conducted a large EMS mutagenesis (Hummel et

al., 1999a; Hummel et al., 1999b). From this screen, we

recovered a large number of zygotically required genes needed

to ensure the correct development of the axon pattern in the

ventral nerve cord in the Drosophila embryo. Here, we present

the analysis of one of these complementation groups that we

initially had called cabrio. We mapped the gene and during the

course of a detailed phenotypic analysis found that cabrio was

allelic to mummy, a complementation group identified by

Nusslein-Volhard and colleagues in a screen for cuticle

differentiation in 1984 (Nusslein-Volhard et al., 1984). As we

show here, mummy is also required for normal CNS midline

glia development and axon fasciculation. Furthermore, we

could show that mummy interferes with dpp expression during

dorsal closure in the embryo as well as in the compound eye.

Molecular cloning of the gene showed that it encodes a well-

conserved UDP-N-acetylglucosamine diphosphorylase, a cen-

tral enzyme in the glycosylation pathway. In mummy mutants,

both the formation of GPI anchors as well as the overall

glycosylation pattern is disrupted. Since Drosophila contains

only one such enzyme, the relatively mild zygotic phenotype

may be explained by either maternal rescue or a still

unidentified UDP-N-acetylglucosamine epimerase that would

salvage the mummy mutant phenotype.

2. Results

2.1. Cabrio Affects axonal pattering in the CNS

Normal formation of the axonal lattice in the Drosophila

nervous system depends in large part on neuron–glia

interaction. In a large-scale phenotypic screen aimed to isolate

mutants affecting the underlying cellular interactions we had

identified four alleles (J1201, Q180, H120, G131) of a

complementation group that we called cabrio (German word

for convertible).

The G131 mutation causes a hypomorphic phenotype,

whereas the Q180, H120 and J1201 mutants correspond to an

amorphic state with a phenotype resembling that of embryos

homozygous for a cabrio deficiency (Fig. 1; data not shown).

Unlike in wild type embryos, where anterior and posterior

commissures can be clearly separated (Fig. 1A), cabrio mutant

embryos do not form discrete anterior and posterior

commissural axon tracts but instead so called fused

commissures (Fig. 1B,F,G). This axonal phenotype is

indicative for defects in the interaction of the midline glial

cells with specific midline neurons (Klambt and Goodman,

1991). To test the differentiation of the midline glial cells we

employed the enhancer trap line P[lacZ] AA142 to follow the

nuclei of the midline glial cells. In contrast to wild type

embryos, slightly fewer midline glial cells are detected that

often are not properly positioned (Figs. 1B, 8B; wild type: on

average three glial cells per neuromere, cabrio mutants: on

average 2.5 glial cells per neuromere. nZ20 stage 16

embryos). Furthermore, the longitudinal connectives appear

thinner and are found closer to the CNS midline (Fig. 1A,B).

Fasciclin II (Fas II) expressing fascicles often cross the midline

and do not fasciculate as tightly as they do in wild type

(Fig. 1C,D, arrowheads). Similarly, we noted fasciculation

defects when we stained for the presence of the Futsch epitope

using 22C10 antibodies (compare Fig. 1E–G, arrwoheads).

Within the peripheral nervous system (PNS), axons do also not

fasciculate as tightly as they do in the wild type. In addition,

neuronal clustering is affected (Fig. 1J). In the wild type PNS,

chordotonal organs are found in a very regular panpipe pattern

whereas their arrangement is irregular in cabrio mutants

(Fig. 1).

2.2. Dorsal closure phenotype of cabrio

In addition to the specific defects during CNS and PNS

development, mutant cabrio embryos also display a ‘dorsal

open’ phenotype due to a characteristic failure to close the

dorsal epidermis. Hence, we chose to name cabrio for the

complementation group. During stage 14 of embryogenesis,

the lateral epidermal cell sheaths move dorsally to cover the

forming midgut and to close the embryo dorsally. In whole

mount preparations of cabrio mutant stage 16 embryos a dorsal

Fig. 1. cabrio is required for embryonic nervous system development. The figure shows dissected CNS (A–G) or PNS (H–J) preparations of stage 16 embryos stained

for BP102 and the AA142 enhancer trap activity (A,B); FasII (C,D); 22C10 (E–J). Anterior is up in A–G, left in H–J, the genotype is indicated. (A) In a wild type

CNS, two segmental commissures connect the right and left halves of the nervous system. At the midline are 3–4 midline glial cells that express the AA142 lineage

marker (blue). Individual neuromeres are connected with connectives. (B) cabrio mutants are characterized by a so called fused commissure phenotype,

the segmental commissures are located closer together and the connectives are closer to the CNS midline. The midline glial cells show a defective migration.

(C) Fasciclin II is expressed in three distinct fascicles within the longitudinal connectives. (D) In cabrio mutants, fasciculation defects are observed. In addition, the

innermost fascicle often crosses the CNS midline. (E) The 22C10 epitope is expressed by a number of discrete fascicles. (F,G) Depending of the allelic strength,

severe fasciculation defects can be observed. (H) Wild type PNS. The sensory neurons express the 22C10 antigen. The PNS axons form distinct fascicles. (I,J) PNS

defects in cabrio mutants. Note the fasciculation defects and the mal-positioned cell bodies of the sensory neurons.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499 489

Fig. 2. mummy affects dorsal closure. (A–C) Cuticle preparations. (D–K) Whole mount preparations stained for the presence of the Fasciclin III antigen. The

genotype is indicated. Anterior is to the left. (A) In wild type embryos a richly differentiated cuticle forms. The ventral denticle belts are clearly recognizable. (B,C)

In mummy mutant larvae we frequently noted a dorsal open phenotype (asterisk). In addition, the larval cuticle is not as differentiated as in wild type larvae. We

noted, however, some variability of the mutant phenotype, with about 50% of the mutants showing no disruption in dorsal closure. (D,E) In wild type embryos,

Fasciclin III is expressed on all cell membranes of the epidermis and the cell shape changes during dorsal closure can be followed. (F,G) In mummyJ1201 mutants

dorsal closure is initiated normally but fails to complete until stage 16. As a consequence, the midgut loops out of the embryo (G, arrow). (H) In a close up view,

elongated cells can be recognized at the leading edge of the dorsal epidermis (arrowheads). (I) In mummyJ1201 mutants, cells of the leading edge are usually of a more

rounded shape (arrowhead). (J) During dorsal closure, epidermal cells line up in a very regular way at the dorsal midline of the embryo (arrow). (K) In those regions

of mummyJ1201 mutants that were able to close the dorsal epidermis, cells stay rounded (arrows). The asterisk indicates the midgut looping out of the embryo. (L)

Sagittal section of a stage 16 wild type embryo stained for Fasciclin III (FasIII) and ß-Spectrin expression. FasIII is present in the lateral domain of epidermal cells

and is never found in basal context of the cell. Cell outlines are labeled using ß-Spectrin which is distributed throughout the baso-lateral membrane domain. (M)

Sagittal section of a stage 16 mummy mutant embryo. FasIII expression extends to the basal cell membrane (arrow). In addition, cells appear slightly larger and more

round in shape. No alteration in the distribution of ß-Spectrin can be detected.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499490

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499 491

hole of varying size is detected (Fig. 2). Whereas, more than

90% of the mutant stage 16 embryos show this defect

(Fig. 2C,F,G), only about 50% of the cabrio mutant larvae

showed a dorsal open phenotype (Fig. 2B,C) suggesting that

some corrections can be made during late embryonic

development. In the cuticle preparations, we also noted a

weak general differentiation of the cuticle (Fig. 2B,C). This is

reminiscent to the phenotype described for mummy mutants

that were identified by Nusslein-Volhard et al. (1984). Since,

our mapping data placed cabrio roughly in the same

chromosomal interval we performed complementation ana-

lyses. All cabrio alleles failed to complement mummy. In order

to avoid confusion, we will use the name mummy from here on.

The gene has also been identified as cystic, (Devine et al.,

2005).

2.3. mummy affects Dpp expression during the dorsal closure

In the wild type, the dorsal most cells of the epidermis show

a typical elongated shape (Fig. 2H,J). These cells move

towards each other during dorsal closure and zipper together in

a very precise pattern (Fig. 2J). In contrast, in mummy mutant

embryos these cells have an abnormal shape. They are

generally more rounded in shape and show an irregular spacing

in the domain of the dorsal epidermis that successfully closes

the embryo (Fig. 2I,K). Cell shape changes are also apparent in

lateral view of epidermal cells. To outline the cell shapes, we

visualized the distribution of ß-Spectrin (Fig. 2L,M, green in

the merge). Furthermore, FasIII expression often extends to the

basal side of the cells, which in mummy mutants have a more

round appearance (Fig. 2K,L).

During Drosophila embryogenesis, the cell shape changes

that initiate dorsal closure are mostly controlled by the Jun-

kinase (JNK) signaling cascade and by dpp signaling (Martin

and Parkhurst, 2004; Xia and Karin, 2004). To test whether

mummy affects dorsal closure via interference of the dpp

expression, we determined dpp RNA distribution in wild type

and mummy mutant embryos. Whereas in wild type embryos

dpp expression is robustly detected in the leading edge of the

moving epidermal cell sheath, we noted a greatly reduced dpp

expression in the leading edge of mummy or jun mutants but no

alterations in other parts of the embryo (Fig. 3). This indicates

that mummy function might be somehow integrated to control

dpp expression. Another target gene of Jun-Kinase signaling in

the leading edge cells is puckered (Martin-Blanco et al., 1998).

However, the expression of puc-lacZ in the leading edge of

stage 14 embryos is only slightly reduced in mummy mutant

embryos (Fig. 3G,H), suggesting that mummy does not

generally affect JNK signaling but may be more specific to

dpp signaling.

dpp Activity activates a positive feedback loop that via JNK

reinforces dpp expression (Glise and Noselli, 1997). To test

whether mummy acts within this genetic circuit we expressed

activated Jun in the dorsal most cells of the ectoderm using the

Gal4 driver lines 69BGal4 and pnrGal4 (see Section 4 for

description of the expression pattern). In both cases, the

mummy mutant phenotype could not be rescued (data not

shown). Thus, mummy is likely to act upstream of dpp in a

pathway parallel to the classical JNK signaling cascade.

Furthermore, mummy appears to function rather pleiotropic

since we found no evidence that Dpp or one of its receptors acts

in the developing CNS midline. Likewise, we could not rescue

the mummy phenotype by expression of activated Jun in the

CNS midline using a simGal4 driver (data not shown).

2.4. mummy affects eye development

The previous result led to the assumption that mummy can in

part regulate Dpp expression. Dpp is not only required during

dorsal closure but is also critically involved in the developing

compound eye (Chanut and Heberlein, 1997; Pignoni and

Zipursky, 1997). Here, dpp is expressed in a narrow stripe of

cells ahead of the morphogenetic furrow of the eye imaginal

disc (Fig. 4A). To determine whether mummy has any effects

on this dpp expression domain we generated mutant cell clones

using the eyFlp/Minute technique. Following induction of large

mutant clones, the adult eyes appeared small and with

irregularly spaced ommatidia (Fig. 4D,F). In the developing

eye, imaginal discs dpp expression appears to be initiated

normally but is rather broad and irregular when compared to

wild type. In addition, ommatidial differentiation is affected

and mature photoreceptors can be found directly adjacent to the

morphogenetic furrow (Fig. 4).

2.5. Identification of the mummy locus

To further understand the molecular basis for the pleiotropic

effects of the mummy mutation, we determined the molecular

nature of the gene. mummy alleles were previously mapped

meiotically to 2–16. Using deficiency mapping, we could place

mummy in the chromosomal region 26D, which is in relatively

good agreement with the meiotic mapping data. Within this

area, the deficiencies Df(2L)BSC6 and Df(2L)ED6461 failed to

complement all mummy alleles identified. The deficiencies

Df(2L)ED6016 and Df(2L)BSC7 complemented mummy

(Fig. 5). To further map mummy within the 20 kb large

genomic interval defined by these deficiencies, we employed a

number of EP-element insertions and performed recombination

mapping and isolated new mummy alleles by excision

mutagenesis. The P-element insertion KG08617 into CG9535

fails to complement all mummy alleles. The CG9535 locus has

two promoters where two transcripts are initiated that encode

two related proteins differing by 37 N-terminal amino acids

(Flybase). The insertion EP2016 element occurs in 5 0 region of

the deduced mummy transcript. The insertion does not lead to a

mutant mummy phenotype and is homozygous viable. A small

deficiency that affects only the first untranslated exon of

mummy was generated by imprecise excision. Homozygous

Df(2L)D2016-3 show the typical mummy mutant phenotype

(Fig. 2C), suggesting that transcription from the first mummy

promoter is crucial for embryonic development. Furthermore,

several EMS induced alleles were sequenced (Devine et al.,

2005); defining CG9535 as the mummy locus (Fig. 5). The

Fig. 3. mummy affects dpp expression. Whole mount preparations stained for the presence of the dpp RNA (A–F) or the activity of the puc-lacZ element (G,H).

Anterior is to the left. RNA in situ hybridization of dpp antisense RNA to wild type (A,B); mummyJ1201 mutants (C,D); or jun1A189 mutants (E,F). (A,B) In wild type

embryos dpp expression is found in the leading edge cells of the dorsal epidermis (arrowhead) and a stripe of cells in the mesoderm. Both cell types express dpp at the

same intensity. In addition, expression of dpp can be seen in a block of cells in the developing midgut. (B–F) In mummy or jun mutant embryos, dpp expression is

unchanged in level in the mesoderm or the midgut, however, dpp in the leading edge cells of the ectoderm is significantly reduced (arrowheads). (G) In wild type

embryos puc-lacZ is expressed in the leading edge cells of the dorsal ectoderm (arrowhead). (H) In mummyJ1201 mutants expression of the puc-lacZ element is

slightly reduced.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499492

same CG9535 was independently identified to correspond to

the cystic locus (Devine et al., 2005).

Digoxygenin labeled CG9535 RNA antisense probes that

recognize both deduced transcripts revealed an early maternal

expression of mummy. Later during development, a low level

of zygotic expression can be detected in almost all tissues.

High level of mummy expression is seen most prominently in

the developing tracheal system. Only low levels of mummy

expression are found in the ectoderm or the developing nervous

system (Fig. 5).

2.6. mummy encodes a UDP transferase

The mummy locus (CG9535) encodes an UDP-N-acetylglu-

cosamine diphosphorylase (EC 2.7.7.23). This enzyme

catalyzes the last step in the biosynthesis of UDP-N-

acetylglucosamine (UDP-GlcNac), the synthesis of UDP-N-

acetylglucosamine from glucosamine-1-phosphate and UTP

(Mio et al., 1998). Eucaryotic UDP-GlcNAc is involved in

chitin synthesis, the generation of glycosyl-phophatidyl

inositol (GPI) anchors and it is the GlcNAc component of

N-linked glycosylation of many extracellular proteins. Droso-

phila mummy is well conserved in eukaryotes. Over 90% of its

open reading frame share about 50% identity with its

mammalian homologues (PZ10K142) and about 40% identity

with its yeast homolog UDP-GlcNac (PZ10K88) BLASTP

(Altschul et al., 1997).

To confirm that UDP-N-acetylglucosamine synthesis is

inhibited in mummy mutant embryos we employed Lectin-

based immunohistochemistry (D’Amico and Jacobs, 1995).

Consistent with the notion that mummy encodes a UDP-N-

acetylglucosamine diphosphorylase, reactivity of FITC-

Fig. 4. mummy affects dpp Expression during eye development. (A–E) dpp Expression is visualized in eye discs using a P[dpp-lacZ] insertion (red). Neurons that

form posterior to the morphogenetic furrow are labeled by green (staining against the Elav protein). Anterior is to the left. (A,C) In wild type eye imaginal discs, dpp

expression is found in a narrow and straight strip of cells close to the morphogenetic furrow. (B,D) mummy mutant eye discs were generated using the eyFlp FRT

Minute technique. In such mutant eye discs dpp expression is highly abnormal. In some regions of the eye discs, dpp expression is missing (arrowheads), whereas in

other parts of the eye disc, broader bands of dpp expression can be seen (arrow). In addition, fewer and irregular spaced photoreceptor neurons form. (E,F) The

disruption of eye imaginal disc development are reflected by the adult eye phenotype. (E) Whereas, wild type flies show regular formed compound eye, mutant

mummy eyes show a severely reduced eye with irregular spacing of ommatidia.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499 493

coupled lectins specific for binding N-acetylglucosamine

oligomers is greatly reduced in the mutant compared to wild

type embryos (Fig. 6). On the other hand, reactivity of lectins

specific for other sugar residues can still be observed (Fig. 6).

Interestingly, staining with PNA, which recognizes galactosyl

(b-1,3) N-acetylgalactosamine revealed ectopic reactivity in

trachea possibly due to the uncovering of epitopes normally

masked by other sugar residues or chitin structures.

To further study whether mummy affects glycosylation, we

first expressed a GPI-anchored GFP protein (Greco et al., 2001)

in a wild type and in a mummy mutant background (Fig. 7).

Whereas normal GFP shows a molecular weight of about

30 kDa, a 40 kDa large GPI-anchored GFP protein can be

detected by anti-GFP antibodies a wild type background

(Fig. 7A). In protein extracts derived from collections

containing 25% mutant mummy embryos anti-GFP antibodies

recognize an additional protein band of about 30 kDa, which

corresponds to the predicted molecular weight of GFP. These

differences in molecular weight could be due to a defective GPI

linkage, which requires UDP-N-acetylglucosamine dipho-

sphorylase activity. However, we detected no gross changes

in the subcellular distribution of GFP in mummy mutants and

Fig. 5. The mummy locus. (A) Schematic representation of the genetic organization of the mummy region. The extent of the deficiencies is indicated. (B) Two

transcripts can be generated from the mummy locus, which are predicted to encode proteins of 520 and 483 amino acids in length. The positions of the different P-

element insertions are indicated. The allele Df(2L)D2016-3 has been isolated following a imprecise excision experiment starting from the EP2016 insertion. (C)

Ubiquitous mummy RNA expression of mummy in a stage 11 embryo. (D) Stage 14 embryo. High levels of mummy expression are found in the developing tracheal

system.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499494

noted only slight changes in the morphology of the epidermal

cells (Fig. 7) that were also noted using anti-FasIII antibodies

(Fig. 2L,M). If GPI-anchor formation had been impaired, we

would have expected that GFP accumulates in the ER (Field et

al., 1994). Since, we detected no increase in the number of such

vesicles we suggest that the observed reduction in the

molecular weight of GFP is due to in glycosylation defects.

To test GPI-linked proteins expressed in the developing

nervous system, we studied Fasciclin II and Wrapper

expression. The adhesion protein Fasciclin II is expressed by

a small subset of axons (Grenningloh et al., 1991) that also

show fasciculation defects in mummy mutants. We could not

detect any changes in the expression pattern or the subcellular

distribution of the Fasciclin II protein, which is expressed as

either GPI-anchored protein or as transmembrane anchored

protein (Grenningloh et al., 1991).

The Wrapper protein is characterized by three immuno-

globulin domains and one fibronectin domain and is

specifically expressed by the midline glial cells. Only one

Fig. 6. Abnormal carbohydrate modifications in mummy mutants. Whole mount pre

embryos (B,D,F,H, lower row) stained for different lectin binding. Anterior is to the

glucosamine. The lectin binding is reduced in mummy mutants when compared to w

conjugated WGA, which has a binding specificity for sialic acid and N-acetyl-gluco

type. (G,H) Binding of FITC-conjugated PNA lectin, which recognizes Galactose. T

by the strong staining in trachea. The tracheal phenotype associated with mummy c

Wrapper isoform has been reported that is attached to the glial

cell membrane by a GPI anchor (Noordermeer et al., 1998).

Expression of Wrapper is clearly abnormal in mutant mummy

embryos (Fig. 8). Whereas, in wild type embryos Wrapper is

exclusively expressed as a GPI-linked protein on the

membrane of the midline glial cells surrounding the segmental

commissures, it is largely localized in intracellular vesicles in

mummy mutant embryos (Fig. 8). Since, wrapper mutants

display a weak commissural phenotype, part of the mummy

mutant phenotype can be explained by mis-localized Wrapper

protein.

3. Discussion

In a genetic screen aimed to identify zygotically active

genes required for CNS development we have found and

characterized the cabrio complementation group, which was

subsequently found to be allelic to the mummy locus. mummy

mutants were initially isolated based on a weakly differentiated

parations of stage 16 wild type (A,C,E,G, upper row) and mummyJ1201 mutant

left. (A–D) Binding of FITC-conjugated DSL lectin that recognizes N-acetyl-

ild type. vnc, ventral nerve cord; ts, tracheal system. (E,F) Staining with FITC-

samine. Again, staining is reduced in mummy mutants when compared to wild

he wild type expression pattern is not abolished in mummy mutants but masked

an be seen.

Fig. 7. Expression of GPI-GFP. (A) Western blot using protein lysates from embryos expressing UAS-GPI-GFP using the da-GAL4 driver. Otherwise wild type

embryos and embryos derived from a collection of heterozygous mummyH120 animals were separated on a 10% PAGE and transferred to nylon membranes.

Antibodies directed against the GFP protein recognize a 40 kDa band corresponding to the GPI-GFP fusion protein. In embryo collections containing 25% mummy

mutants an additional 28 kDa protein band is detected. (B,C) Epidermis of a stage 16 embryo expressing expressing GPI-anchored GFP stained for GFP (green) and

Discs large (red) expression. Discs large is a cytoplasmic protein closely associated with the cell membrane. (B) Wild type embryo. GFP is found mostly at the cell

membrane only few vesicular structures are shown. (C) mummyJ1201 mutant embryo. No differences to the wild type can be detected in the GFP expression.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499 495

cuticle but as we show here have in addition defects in dorsal

closure, dpp expression in the ectoderm and the eye imaginal

disc as well as defects in the development of the central and

peripheral nervous system.

mummy encodes an UDP-N-acetylglucosamine diphosphor-

ylase that catalyses the formation of UDP-N-acetylglcosamine

(UDP-GlcNac) from UTP and N-acetylglucosamin-1-phos-

phate (Araujo et al., 2005; Devine et al., 2005; Tonning et al.,

2006). The protein is well conserved and it was recently shown

that the human Mummy homolog is capable to rescue the yeast

UDP-N-acetylglucosamine diphosphorylase mutant (Mio et al.,

1998). UDP-GlcNac is a central and ubiquitous metabolite that

is required in many different reactions. In invertebrates, UDP-

GlcNac is the substrate of the chitin synthase and thus is

required for the formation of the exoskeleton (Araujo et al.,

2005; Devine et al., 2005; Merzendorfer and Zimoch, 2003;

Tonning et al., 2006). This requirement actually led to the first

identification of mummy alleles, as these mutants failed to

generate a fully differentiated cuticle but showed no overall

patterning defects (Nusslein-Volhard et al., 1984). The

disruption of chitin synthesis also caused tracheal tube

phenotypes and tube size has been subsequently linked to

chitin synthesis (Devine et al., 2005; Tonning et al., 2005).

Several genes involved in chitin synthesis are known:

knickkopf (knk) and krotzkopf verkehrt (kkv) (Moussian et al.,

2005; Ostrowski et al., 2002). Whereas kkv encodes a chitin

synthase knk encodes a novel protein that may be involved in

either targeting or secretion of chitin or chitin synthesizing

enzymes (Devine et al., 2005; Moussian et al., 2006; Ostrowski

et al., 2002). To determine whether mummy mutants show

nervous system phenotypes due to a so far unnoticed role of

chitin during CNS development, we determined the nervous

system phenotype of both knk and kkv but did not detect any

abnormal neural phenotypes. Therefore, we conclude that

mummy functions in a different pathway during neuronal

development.

The UDP-GlcNac generated by Mummy is not only used in

chitin synthesis. It is also used during N-linked glycosylation

and the formation of the glycosyl-phosphatidyl-inositol (GPI)

anchor that keeps some extracellular proteins attached to the

outer leaflet of the plasma membrane (Eisenhaber et al., 2003).

In wild type embryos, the GPI-anchor is generated in the ER

and the modified protein is then transported to the plasma

membrane. Following disruption of GPI-anchor formation,

proteins are expected to be accumulating in the ER (Eisenhaber

et al., 2003; Field et al., 1994; Vidugiriene and Menon, 1994).

To test this hypothesis, we analyzed the expression of two GPI-

anchored proteins: GPI-GFP and GPI-Wrapper. We could not

observe any intracellular accumulation of the GPI-anchored

GFP protein in epidermal cells of mummy mutants, suggesting

that maternal function is sufficient at least in epidermal cells.

We noted a similar maternal contribution when analyzing pig-T

mutants, where the transfer of the GPI-anchor to substrate

proteins is affected (Bourbon et al., 2002; Eisenhaber et al.,

2003). However, we did note a strong intracellular accumu-

lation of the Wrapper protein in the midline glial cells of the

embryonic nervous system. wrapper encodes a 500 amino

acids large protein carrying three immunoglobulin domains

that is specifically expressed as a GPI-linked protein on the

surface of the midline glia (Noordermeer et al., 1998). In

Fig. 8. Expression of GPI-Wrapper in mummy mutants. Dissected CNS preparations of stage 16 embryos stained for the presence of the HRP epitope (blue) to mark

neurons and the GPI-anchored Wrapper protein (red) expressed by the midline glial cells. Single confocal sections are shown. Anterior is up. (A) In wild type

embryos the midline glial cells ensheath the segmental commissures. The outlines of the midline glial cells express the Wrapper protein (red). The dotted area is

shown in higher magnification in (B). (C) In mummyH1-20 mutant embryos, the midline glial cells do not correctly wrap the segmental commissures and the

commissures appear fused. Wrapper expression is no longer confined to the cell membrane but found in vesicular structures accumulating within the glial cells

(arrowhead). No clear staining of the cell membrane can be detected. The dotted area is shown in higher magnification in (D).

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499496

mummy mutants, the Wrapper protein is found to a large extent

in intracellular vesicles, which are never observed in wild type

midline glial cells. Thus, only reduced levels of Wrapper reach

the cell surface, which contributes at least in part to the

embryonic mummy CNS phenotype. wrapper mutant embryos

are characterized by an incomplete wrapping of the segmental

commissures (Noordermeer et al., 1998). Similarly, differen-

tiation of the midline glia is impaired in mummy mutants, albeit

to a much stronger extent.

In addition to the defects in GPI anchor formation,

mummy mutants display an abnormal dpp expression in the

embryo as well as during eye disc development. Given the

well-documented interactions of signaling molecules such as

Dpp, Hh and Wg, with heparan sulfate proteoglycans

(Belenkaya et al., 2004; Hacker et al., 2005; Han et al.,

2005; Lee and Chien, 2004; Lin, 2004) these phenotypes are

likely to be explained by glycosylation defects. During

embryogenesis disrupted Dpp diffusion may interfere with a

positive feed-back loop controlling dpp expression (Ashe,

2005; Glise and Noselli, 1997), whereas, during eye

development, irregular dpp expression may also reflect

impaired Hedgehog signaling due to disrupted interactions

with the extracellular matrix (Callejo et al., 2006; Perrimon

and Hacker, 2004; The et al., 1999).

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499 497

Our finding that several lectins do not recognize specific

carbohydrate structures in mummy mutants supports the notion

that glycosylation is globally affected (Fig. 7, see also (Araujo

et al., 2005)). Two previously identified mutants, nac and mas-

1 also affect glycosylation but show only relatively mild

defects during nervous system development and show no

alterations in the overall morphogenetic processes as we noted

them for mummy mutants (Kerscher et al., 1995; Whitlock,

1993). It had been suggested that a complete genetic block in

glycoprotein processing might lead to gross axon guidance

phenotypes (Kerscher et al., 1995; Lee and Chien, 2004). Our

phenotypic analyses of mummy mutants are in agreement with

this notion, however, since the phenotype is relatively mild,

they also suggest that in addition to a strong maternal

component there might be an unidentified salvage pathway

that can at least provide some UDP-N-acetylglucosamine to the

animal. In summary, the phenotype of zygotic mummy mutants

appears to be due to the combination of defects in GPI-

anchorage and defects on the glycosylation of many proteins.

4. Materials

4.1. Genetics

The following fly stocks were used: mummyZ234; mummyZ237; EP2016;

KG8617; KG4349; UAS-GPI-GFP (kindly provided by S. Eaton); Df(2L)E110;

Df(2L)BSC6; Df(2L)ED6461, (kindly provided by G. Reuter); Df(2L)ED6016;

Df(2L)BSC7. The 69BGal4 (directs expression in the embryonic epidermis) and

pnrGal4 (directs expression in the dorsal epidermis) in the pannier pattern,

simGal directs expression in all CNS midline cells (Menne et al., 1997). All fly

work was done according common practice. The Df(2L)D2016 was generated

using a standard excision mutagenesis protocol starting from the homozygous

viable P-element insertion EP(2)2016. mummyJ1201 FRT40A recombinant

chromosomes were used for the generation of homozygous cell clones using an

ey-Flp source (Newsome et al., 2000). The use of a cell lethal leads to eye discs

containing to about 90–100% mutant cells (Newsome et al., 2000). All stocks

were obtained from the Bloomington stock center if not otherwise indicated.

4.2. Immunhistohemistry

Preparation of embryos and imaginal discs with subsequent staining for the

expression of the different antigens was performed as described previously

(Schimmelpfeng et al., 2001). Cuticle preparations were done according to

(Nusslein-Volhard et al., 1984). The following antibodies were obtained from

the Developmental Studies Hybridoma Bank at the University of Iowa: anti-

Futsch: 22C10 (Hummel et al., 2000); BP102 (A. Bieber, N. Patel, C.S.

Goodman, unpublished); anti-FasII: 1D4 (C.S. Goodman, unpublished); anti-

FasIII: 7G10 (Snow et al., 1989), anti-Wrapper (Noordermeer et al., 1998).

Cofocal images were taken on a Zeiss LSM 510 meta. Anti-HRP and secondary

antibodies were obtained from Dianova (Hamburg, Germany). In situ

hybridization to RNA was performed as described (Tautz and Pfeifle, 1989).

4.3. Lectin immunofluorescence

Eight to sixteen hours old Drosophila embryos were collected from apple

juice agar plates, dechorionated in 5% commercial bleach, rinsed in water and

then fixed in n-heptane previously saturated with 4% formaldehyde in

phosphate-buffered saline (PBS) for 30 min. After fixation, the vitelline

membrane was removed by cracking at a heptane/methanol interphase followed

by rehydration and washing in PBT (PBS, 0.3% Triton-X100). Fluorescein-

coupled lectins were employed at a concentration of 50 mg/ml over night at

4 8C. Imaging was performed using an Olympus FV1000 confocal microscope.

The following Fluorescein labeled lectins were obtained from the Vector

Laboratories: Datura Stramonium Lectin (DSL) and Wheat Germ Agglutinin

(WGA) that bind N-acetylglucosamine (GlcNac) oligomers and Peanut

Agglutinin (PNA) that binds galactosyl (b-1,3) N-acetylgalactosamine.

Acknowledgements

We thank M. Haenlin, G. Reuter, S. Eaton and the

Bloomington stock center for sending flies. M. Krasnow, B.

Moussian for communication of results prior publication, P.

Grewal and J. Marth for reagents and A. Mertens and J.

Veerkamp for help. This work was supported through a grant of

the DFG to C.K. (SFB492).

References

Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z.,

Miller, W., Lipman, D.J., 1997. Gapped BLAST and PSI-BLAST: a

new generation of protein database search programs. Nucleic Acids Res.

25, 3389–3402.

Araujo, S.J., Aslam, H., Tear, G., Casanova, J., 2005 mummy/cystic encodes an

enzyme required for chitin and glycan synthesis, involved in trachea,

embryonic citicle and CNS development—Analysis of its role in

Drosophila tracheal morphogenesis. Dev Biol 288 (1): 179–93.

Ashe, H.L., 2005. BMP signalling: synergy and feedback create a step gradient.

Curr. Biol. 15, R375–R377.

Barallobre, M.J., Pascual, M., Del Rio, J.A., Soriano, E., 2005. The netrin

family of guidance factors: emphasis on netrin-1 signalling. Brain Res.

Brain Res. Rev. 49, 22–47.

Belenkaya, T.Y., Han, C., Yan, D., Opoka, R.J., Khodoun, M., Liu, H., Lin, X.,

2004. Drosophila Dpp morphogen movement is independent of dynamin-

mediated endocytosis but regulated by the glypican members of heparan

sulfate proteoglycans. Cell 119, 231–244.

Bornemann, D.J., Duncan, J.E., Staatz, W., Selleck, S., Warrior, R., 2004.

Abrogation of heparan sulfate synthesis in Drosophila disrupts the

wingless, hedgehog and decapentaplegic signaling pathways. Development

131, 1927–1938.

Bourbon, H.M., Gonzy-Treboul, G., Peronnet, F., Alin, M.F., Ardourel, C.,

Benassayag, C., et al., 2002. A P-insertion screen identifying novel

X-linked essential genes in Drosophila. Mech. Dev. 110, 71–83.

Bruckner, K., Perez, L., Clausen, H., Cohen, S., 2000. Glycosyltransferase

activity of fringe modulates Notch-Delta interactions. Nature 406, 411–415.

Bulow, H.E., Hobert, O., 2004. Differential sulfations and epimerization define

heparan sulfate specificity in nervous system development. Neuron 41,

723–736.

Bulow, H.E., Berry, K.L., Topper, L.H., Peles, E., Hobert, O., 2002. Heparan

sulfate proteoglycan-dependent induction of axon branching and axon

misrouting by the Kallmann syndrome gene kal-1. Proc. Natl Acad. Sci.

USA 99, 6346–6351.

Callejo, A., Torroja, C., Quijada, L., Guerrero, I., 2006. Hedgehog lipid

modifications are required for hedgehog stabilization in the extracellular

matrix. Development 133, 471–483.

Chanut, F., Heberlein, U., 1997. Role of decapentaplegic in initiation and

progression of the morphogenetic furrow in the developing Drosophila

retina. Development 124, 559–567.

D’Amico, P., Jacobs, J.R., 1995. Lectin histochemistry of the Drosophila

embryo. Tissue Cell 27, 23–30.

Desai, C.J., Popova, E., Zinn, K., 1994. A Drosophila receptor tyrosine

phosphatase expressed in the embryonic CNS and larval optic lobes is a

member of the set of proteins bearing the “HRP” carbohydrate epitope.

J. Neurosci. 14, 7272–7283.

Devine, W.P., Lubarsky, B., Shaw, K., Luschnig, S., Messina, L., Krasnow,

M.A., (2005). Requirement for chitin biosynthesis in epithelial tube

morphogenesis. Proc. Natl Acad Sci U.S.A. 102(47): 17014–9.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499498

Dickson, B.J., 2002. Molecular mechanisms of axon guidance. Science 298,

1959–1964.

Doherty, P., Cohen, J., Walsh, F.S., 1990. Neurite outgrowth in response to

transfected N-CAM changes during development and is modulated by

polysialic acid. Neuron 5, 209–219.

Eisenhaber, B., Maurer-Stroh, S., Novatchkova, M., Schneider, G.,

Eisenhaber, F., 2003. Enzymes and auxiliary factors for GPI lipid anchor

biosynthesis and post-translational transfer to proteins. Bioessays 25, 367–

385.

Field, M.C., Moran, P., Li, W., Keller, G.A., Caras, I.W., 1994. Retention and

degradation of proteins containing an uncleaved glycosylphosphatidylino-

sitol signal. J. Biol. Chem. 269, 10830–10837.

Glise, B., Noselli, S., 1997. Coupling of Jun amino-terminal kinase and

decapentaplegic signaling pathways in Drosophila morphogenesis. Genes

Dev. 11, 1738–1747.

Greco, V., Hannus, M., Eaton, S., 2001. Argosomes: a potential vehicle for the

spread of morphogens through epithelia. Cell 106, 633–645.

Grenningloh, G., Rehm, E.J., Goodman, C.S., 1991. Genetic analysis of growth

cone guidance in Drosophila: fasciclin II functions as a neuronal

recognition molecule. Cell 67, 45–57.

Hacker, U., Nybakken, K., Perrimon, N., 2005. Heparan sulphate proteogly-

cans: the sweet side of development. Nat. Rev. Mol. Cell. Biol. 6, 530–541.

Haines, N., Irvine, K.D., 2003. Glycosylation regulates Notch signalling. Nat.

Rev. Mol. Cell Biol. 4, 786–797.

Haltiwanger, R.S., Stanley, P., 2002. Modulation of receptor signaling by

glycosylation: fringe is an O-fucose-beta1,3-N-acetylglucosaminyltransfer-

ase. Biochim. Biophys. Acta 1573, 328–335.

Han, C., Belenkaya, T.Y., Khodoun, M., Tauchi, M., Lin, X., 2004. Distinct and

collaborative roles of Drosophila EXT family proteins in morphogen

signalling and gradient formation. Development 131, 1563–1575.

Han, C., Yan, D., Belenkaya, T.Y., Lin, X., 2005. Drosophila glypicans dally

and dally-like shape the extracellular wingless morphogen gradient in the

wing disc. Development 132, 667–679.

Herscovics, A., Orlean, P., 1993. Glycoprotein biosynthesis in yeast. Faseb.

J. 7, 540–550.

Holt, C.E., Dickson, B.J., 2005. Sugar codes for axons? Neuron 46, 169–172.

Hummel, T., Schimmelpfeng, K., Klambt, C., 1999a. Commissure formation in

the embryonic CNS of Drosophila: I Identification of the required gene

functions. Dev. Biol. 209, 381–398.

Hummel, T., Schimmelpfeng, K., Klambt, C., 1999b. Commissure formation in

the embryonic CNS of Drosophila: II function of the different midline cells.

Development 126, 771–779.

Hummel, T., Krukkert, K., Roos, J., Davis, G., Klambt, C., 2000. Drosophila

futsch/22C10 is a MAP1B-like protein required for dendritic and axonal

development. Neuron 26, 357–370.

Johnson, K.G., Ghose, A., Epstein, E., Lincecum, J., O’Connor, M.B., Van

Vactor, D., 2004. Axonal heparan sulfate proteoglycans regulate the

distribution and efficiency of the repellent slit during midline axon

guidance. Curr. Biol. 14, 499–504.

Kerscher, S., Albert, S., Wucherpfennig, D., Heisenberg, M., Schneuwly, S.,

1995. Molecular and genetic analysis of the Drosophila mas-1 (mannosi-

dase-1) gene which encodes a glycoprotein processing alpha 1,2-

mannosidase. Dev. Biol. 168, 613–626.

Kirkpatrick, C.A., Dimitroff, B.D., Rawson, J.M., Selleck, S.B., 2004. Spatial

regulation of Wingless morphogen distribution and signaling by dally-like

protein. Dev. Cell 7, 513–523.

Klambt, C., Goodman, C.S., 1991. The diversity and pattern of glia during axon

pathway formation in the Drosophila embryo. Glia 4, 205–213.

Lee, J.S., Chien, C.B., 2004. When sugars guide axons: insights from heparan

sulphate proteoglycan mutants. Nat. Rev. Genet. 5, 923–935.

Lee, J.S., von der Hardt, S., Rusch, M.A., Stringer, S.E., Stickney, H.L.,

Talbot, W.S., et al., 2004. Axon sorting in the optic tract requires HSPG

synthesis by ext2 (dackel) and extl3 (boxer). Neuron 44, 947–960.

Lin, X., 2004. Functions of heparan sulfate proteoglycans in cell signaling

during development. Development 131, 6009–6021.

Martin, P., Parkhurst, S.M., 2004. Parallels between tissue repair and embryo

morphogenesis. Development 131, 3021–3034.

Martin-Blanco, E., Gampel, A., Ring, J., Virdee, K., Kirov, N.,

Tolkovsky, A.M., Martinez-Arias, A., 1998. Puckered encodes a

phosphatase that mediates a feedback loop regulating JNK activity during

dorsal closure in Drosophila. Genes Dev. 12, 557–570.

Menne, T.V., Luer, K., Technau, G.M., Klambt, C., 1997. CNS midline cells in

Drosophila induce the differentiation of lateral neural cells. Development

124, 4949–4958.

Merzendorfer, H., Zimoch, L., 2003. Chitin metabolism in insects: structure,

function and regulation of chitin synthases and chitinases. J. Exp. Biol. 206,

4393–4412.

Mio, T., Yabe, T., Arisawa, M., Yamada-Okabe, H., 1998. The eukaryotic

UDP-N-acetylglucosamine pyrophosphorylases. Gene cloning, protein

expression, and catalytic mechanism. J. Biol. Chem. 273, 14392–14397.

Moloney, D.J., Panin, V.M., Johnston, S.H., Chen, J., Shao, L., Wilson, R.,

et al., 2000. Fringe is a glycosyltransferase that modifies Notch. Nature 406,

369–375.

Moussian, B., Schwarz, H., Bartoszewski, S., Nusslein-Volhard, C., 2005.

Involvement of chitin in exoskeleton morphogenesis in Drosophila

melanogaster. J. Morphol. 264, 117–130.

Moussian, B., Tang, E., Tonning, A., Helms, S., Schwarz, H., Nusslein-

Volhard, C., Uv, A.E., 2006. Drosophila knickkopf and retroactive are

needed for epithelial tube growth and cuticle differentiation through their

specific requirement for chitin filament organization. Development 133,

163–171.

Nakato, H., Fox, B., Selleck, S.B., 2002. Dally, a Drosophila member of the

glypican family of integral membrane proteoglycans, affects cell cycle

progression and morphogenesis via a cyclin a-mediated process. J. Cell Sci.

115, 123–130.

Newsome, T.P., Asling, B., Dickson, B.J., 2000. Analysis of Drosophila

photoreceptor axon guidance in eye-specific mosaics. Development 127,

851–860.

Noordermeer, J.N., Kopczynski, C.C., Fetter, R.D., Bland, K.S., Chen, W.Y.,

Goodman, C.S., 1998. Wrapper, a novel member of the Ig superfamily, is

expressed by midline glia and is required for them to ensheath commissural

axons in Drosophila. Neuron 21, 991–1001.

Nusslein-Volhard, C., Wieschaus, E., Kluding, H., 1984. Mutations affecting

the pattern of the larval cuticle in Drosophila melanogaster. I. Zygotic loci

on the second chromosome. Roux’s. Arch. Dev. Biol. 193, 267–282.

Nybakken, K., Perrimon, N., 2002. Heparan sulfate proteoglycan modulation of

developmental signaling in Drosophila. Biochim. Biophys. Acta 1573,

280–291.

Okajima, T., Xu, A., Irvine, K.D., 2003. Modulation of notch-ligand binding by

protein O-fucosyltransferase 1 and fringe. J. Biol. Chem. 278, 42340–

42345.

Ostrowski, S., Dierick, H.A., Bejsovec, A., 2002. Genetic control of cuticle

formation during embryonic development of Drosophila melanogaster.

Genetics 161, 171–182.

Perrimon, N., Hacker, U., 2004. Wingless, hedgehog and heparan sulfate

proteoglycans. Development 131, 2509–2511 (author reply 2511–3).

Pignoni, F., Zipursky, S.L., 1997. Induction of Drosophila eye development by

decapentaplegic. Development 124, 271–278.

Sasamura, T., Sasaki, N., Miyashita, F., Nakao, S., Ishikawa, H.O., Ito, M.,

et al., 2003. Neurotic, a novel maternal neurogenic gene, encodes an

O-fucosyltransferase that is essential for Notch-Delta interactions.

Development 130, 4785–4795.

Schimmelpfeng, K., Gogel, S., Klambt, C., 2001. The function of leak and

kuzbanian during growth cone and cell migration. Mech. Dev. 106, 25–36.

Snow, P.M., Patel, N.H., Harrelson, A.L., Goodman, C.S., 1987. Neural-

specific carbohydrate moiety shared by many surface glycoproteins in

Drosophila and grasshopper embryos. J. Neurosci. 7, 4137–4144.

Snow, P.M., Bieber, A.J., Goodman, C.S., 1989. Fasciclin III: a novel

homophilic adhesion molecule in Drosophila. Cell 59, 313–323.

Steigemann, P., Molitor, A., Fellert, S., Jackle, H., Vorbruggen, G., 2004.

Heparan sulfate proteoglycan syndecan promotes axonal and myotube

guidance by slit/robo signaling. Curr. Biol. 14, 225–230.

Sun, B., Salvaterra, P.M., 1995. Characterization of nervana, a Drosophila

melanogaster neuron-specific glycoprotein antigen recognized by anti-

horseradish peroxidase antibodies. J. Neurochem. 65, 434–443.

K. Schimmelpfeng et al. / Mechanisms of Development 123 (2006) 487–499 499

Takei, Y., Ozawa, Y., Sato, M., Watanabe, A., Tabata, T., 2004. Three

Drosophila EXT genes shape morphogen gradients through synthesis of

heparan sulfate proteoglycans. Development 131, 73–82.

Tautz, D., Pfeifle, C., 1989. A non-radioactive in situ hybridization method for

the localization of specific RNAs in Drosophila embryos reveals

translational control of the segmentation gene hunchback. Chromosoma

98, 81–85.

The, I., Bellaiche, Y., Perrimon, N., 1999. Hedgehog movement is regulated

through tout velu-dependent synthesis of a heparan sulfate proteoglycan.

Mol. Cell 4, 633–639.

Tonning, A., Hemphala, J., Tang, E., Nannmark, U., Samakovlis, C.,

Uv, A., 2005. A transient luminal chitinous matrix is required to model

epithelial tube diameter in the Drosophila trachea. Dev. Cell 9,

423–430.

Tonning, A., Helms, S., Schwarz, H., Uv, A.E., Moussian, B., 2006. Hormonal

regulation of mummy is needed for apical extracellular matrix formation

and epithelial morphogenesis in Drosophila. Development 133, 331–341.

Vidugiriene, J., Menon, A.K., 1994. The GPI anchor of cell-surface proteins is

synthesized on the cytoplasmic face of the endoplasmic reticulum. J. Cell

Biol. 127, 333–341.

Whitlock, K.E., 1993. Development of Drosophila wing sensory neurons in

mutants with missing or modified cell surface molecules. Development

117, 1251–1260.

Wong, K., Park, H.T., Wu, J.Y., Rao, Y., 2002. Slit proteins: molecular

guidance cues for cells ranging from neurons to leukocytes. Curr. Opin.

Genet. Dev. 12, 583–591.

Xia, Y., Karin, M., 2004. The control of cell motility and epithelial

morphogenesis by jun kinases. Trends Cell Biol. 14, 94–101.