Access to this full-text is provided by Springer Nature.
Content available from Scientific Reports
This content is subject to copyright. Terms and conditions apply.
1
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
Inter-axonal recognition organizes
Drosophila olfactory map formation
Gaurav Goyal1, Ariane Zierau2, Marc Lattemann2, Beate Bergkirchner1, Dominik Javorski1,
Rashmit Kaur
1 & Thomas Hummel1,2
Olfactory systems across the animal kingdom show astonishing similarities in their morphological and
functional organization. In mouse and Drosophila, olfactory sensory neurons are characterized by the
selective expression of a single odorant receptor (OR) type and by the OR class-specic connection in
the olfactory brain center. Monospecic OR expression in mouse provides each sensory neuron with a
unique recognition identity underlying class-specic axon sorting into synaptic glomeruli. Here we show
that in Drosophila, although OR genes are not involved in sensory neuron connectivity, aerent sorting
via OR class-specic recognition denes a central mechanism of odortopic map formation. Sensory
neurons mutant for the Ig-domain receptor Dscam converge into ectopic glomeruli with single OR class
identity independent of their target cells. Mosaic analysis showed that Dscam prevents premature
recognition among sensory axons of the same OR class. Single Dscam isoform expression in projecting
axons revealed the importance of Dscam diversity for spatially restricted glomerular convergence.
These data support a model in which the precise temporal-spatial regulation of Dscam activity controls
class-specic axon sorting thereby indicating convergent evolution of olfactory map formation via self-
patterning of sensory neurons.
In mouse and Drosophila, olfactory receptor neurons (ORNs) express a single odorant receptor and all neurons of
the same OR class converge into distinct synaptic glomeruli1. However, dierent developmental control mecha-
nisms seem to be employed in the formation of these olfactory maps. In mammals, odorant receptors are critical
determinants in ORN connectivity by mediating inter-axonal communication2–5. A current model proposes that
ORs are not directly involved in axon-axon interaction but that OR endogenous activity leads to the expression
of a distinct set of cell adhesion molecules6,7. is ORN class-specic adhesion code determines local axon sort-
ing and glomerular convergence at a dened position in the olfactory bulb6,7. As the postsynaptic target cells are
largely dispensable for class-specic ORN axon sorting8–10, the formation of an odortopic map in the mouse brain
is thought to be controlled by the self-organizing activity of the sensory neurons via the precise regulation of
unique axonal recognition identities11,12.
In Drosophila, the 50 ORN classes in the adult olfactory system show a similar level of sensory and synaptic
specicity as in mammals13. Each sensory neuron expresses only one type of OR and all neurons of the same OR
class converge their axons onto a single synaptic target unit in the brain14–17. Inside each of these synaptic glo-
meruli, ORN axons associate with two main classes of CNS target dendrites, a glomerulus-specic type of relay
projection neurons (PNs) and various classes of multi-glomerular local interneurons that mediate signal inte-
gration within the rst olfactory processing center18–20. Despite the similarities in olfactory system organization,
OR deletion and misexpression experiments indicate that the Drosophila ORs do not have a function in either
monospecic receptor expression or the determination of synaptic identity21–24. is raises the question if olfac-
tory map formation in Drosophila is also organized by an inter-axonal signaling process similar to mammals or if
direct axon-dendrite interaction with the target eld is the primary patterning mechanism. Previous studies have
shown that glomerulus-specic projection neurons (PNs) pre-pattern the target region before aerent arrival
suggesting olfactory circuitry formation via direct synaptic partner recognition13, though recent studies indicate
a more prominent role of interactions between targeting axons in patterning the antennal lobes25.
Using genetic approaches, Drosophila mutant analyses have identied several cell surface molecules essential
for distinct steps in ORN axon targeting26–30. e most promising candidate to provide unique recognition iden-
tities is the Immunoglobulin (Ig) domain receptor Dscam31,32. Alternative splicing of Dscam can generate more
than 18,000 distinct recognition molecules that bind in a strictly homophilic manner33–35. In nervous system
1Department for Neurobiology, University of Vienna, Althanstrasse 14, 1090, Vienna, Austria. 2Institut für Neuro- und
Verhaltensbiologie, Universität Münster, Badestr. 9, D-48149, Münster, Germany. Correspondence and requests for
materials should be addressed to G.G. (email: gaurav.goyal@univie.ac.at)
Received: 29 May 2018
Accepted: 26 July 2019
Published: xx xx xxxx
OPEN
Content courtesy of Springer Nature, terms of use apply. Rights reserved
2
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
development, Dscam controls the elaboration of axonal and dendritic processes in a cell-autonomous fashion
called self-avoidance36–46. In the same context, Dscam diversity is required to prevent repulsion between neigh-
boring neuronal processes (non-self association)36,40,41. However, no specicity in isoform expression seems to
be necessary for neuronal patterning as randomly selected isoforms can substitute for the loss of the endogenous
Dscam function36,37,39–41,46.
We have shown before that the absence of Dscam causes ORN axons to mistarget into ectopic glomeruli26
raising the question of the underlying cellular interactions. Here we demonstrate that aerent sorting via OR
class-specic recognition denes a critical mechanism for Drosophila olfactory map formation. Loss of Dscam
induces premature inter-axonal recognition with ectopic convergence independent of synaptic partner neurons.
Our experiments suggest that Dscam, although dispensable for inter-axonal recognition itself, is critical for the
precise spatio-temporal regulation of this recognition process. us, despite a dierent requirement of odorant
receptors, olfactory map formation in Drosophila and mouse is mediated by similar mechanisms of aerent sort-
ing via OR-class specic recognition identities.
Results
Dscam mutant ORN axons maintain their class-specific recognition identity. In the adult
Drosophila olfactory system, about 1500 ORNs located in peripheral epithelia subdivide into 50 functional classes
according to the OR expression13. Each sensory class consistsof an average of 20–30 ORNs which converge
their axons to a single synaptic glomerulus on the ipsi-lateral antennal lobe (AL; Fig.1A,C,E)13. Due to this OR
monospecicity, each of the 50 glomeruli in the antennal lobe contains axons of only a single OR class thereby
providing the morphological basis of odor recognition13. To understand the role of Dscam in OR monospeci-
city, we generated Dscam mutant clones specically in ORN precursors of the antennal disc using eyless-p47.
To increase the number of homozygous mutant ORNs, the Dscam+ chromosome contained a cell lethal PCNA
allele48 (Methods). In these ORN specic Dscam mosaics, ORN axons project into additional glomerulus-like
structures26 (Fig.1B,D,F). is mutant axon coalescence can be in close neighborhood to the cognate target glo-
merulus (“glomerular split”, e.g. Fig.1D,F) or more distant to the target region inside or even outside the antennal
lobe (“ectopic glomeruli”, e.g. Fig.1B).
We rst addressed the question if these additional glomeruli maintain aerent OR monospecicity or contain
axons of multiple ORN classes. Local convergence defects were analyzed by visualizing the projection of ORN
class pairs that target two neighboring glomeruli (Fig.1D,F, Supplementary Fig.S1). In all ORN specic Dscam
mosaics with large mutant clones that show a local rearrangement of the glomerular position (ORN47a/22a
n = 14; ORN47b/88a n = 22), axons of dierent ORN classes do not intermingle but segregate according to the
OR identity. To determine the OR class identity of distant ectopic glomeruli we combined a broad ORN marker
and a non-overlapping single OR class marker (Supplementary Fig.S2). In all of the analyzed ORN specic
Dscam mutant mosaic brains (Con > CD2/OR:sytGFP, n > 10 per OR class; MT14 > CD2/OR:sytGFP, n = 10 per
OR class) ectopic glomeruli in dierent regions of the antennal lobe target area maintain their single ORN-class
identity. From these data we conclude that Dscam mutant ORN axons are still able to sort out according to their
sensory class identity.
Class-specic axon coalescence is independent of target neurons. To determine if OR class spe-
cic axon sorting is mediated through a direct interaction with their target cells we analyzed Dscam mutant
maxillary ORNs, which stop frequently outside the AL target area into multiple glomerulus-like structures26
(Fig.1G–J). ese clones were again generated using ORN specic eyless-p with cell lethal PCNA allele on
Dscam+ chromosome to generate large clones (Methods). First, we combined a general marker line including all
of the six maxillary ORN classes together with an ORN class-specic marker (Fig.1G). In Dscam ORN mosaic
brains, we frequently observed a cluster of up to six ectopic glomeruli ventrally to the AL, in which all ORN
axons of the same OR identity are conned to a single glomerulus (n = 15, Fig.1H). Second, we labeled dierent
pairs of maxillary ORN classes, which in wild type project into neighboring or more distant glomeruli ((Fig.1I,
Supplementary Fig.S3). Similar to ectopic coalescence inside the AL, Dscam mutant ORN axons that coalesce
outside the AL target area segregate strictly according to their OR class identity (Fig.1J, n > 30, Supplementary
Fig.S3). As these ectopic glomeruli do not contain dendrites of projection neurons (PNs) (see below), we con-
clude that direct inter-axonal signaling is a critical component of ORN axons sorting.
Local interneurons but not projection neurons innervate ectopic glomeruli. In wild type
Drosophila, the class-specic coalescence of ORN axons into distinct glomeruli is matched by a similar level of
class-specic PN dendrite innervation18,49. As Dscam does not mediate ORN recognition identity, we asked if it
might be involved in axon-dendrite matching. To test this, we generated ORN specic large Dscam mutant clones
using eyless-p/PCNA, thus PNs and LNs were Dscam+/−. In wild type (no Dscam mutation in the background),
ORN88a axons connect to Mz19-positive PNs50. e removal of Dscam from ORNs leads to the local reorgani-
zation of the glomerular eld, in which ORN88a glomeruli are oen displaced by neighboring ORN47b axons
(compare Fig.1E,F). In all analyzed Dscam mosaic brains (n > 10), the changes in glomerulus localization lead to
a corresponding shi in the PN dendritic eld, ensuring that the class-specic ORN-PN matching is maintained
(Fig.2B–E). For more distant ectopic glomeruli we characterized the innervation of Dscam mutant ORN axons
with GH146-positive PN dendrites, which cover all regions of the AL. e ORN classes 21a and 47a, which are
not innervated by GH146-positive PNs in wild type, form ectopic glomeruli in Dscam mosaics but do not receive
GH146-innervation (Fig.2F,H, Supplementary Fig.S4; n > 40). In addition, the Dscam mutant maxillary ORN
46a that converge outside the AL are never associated with GH146-positive PN dendrites although in wild type
this ORN class connects to GH146-PNs (Fig.2G,J; n = 6).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
3
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Figure 1. Dscam mutant axons converge class-specically in ectopic spots. (A,B) Wild type ORN47a axons
(red, anti-CD2) grow to their specic target and converge into one glomerulus (green, sytGFP) (A) whereas
Dscam mutant axons converge into an ectopic glomerular-like structure (B). (C–J) Double labelling of
dierent ORN classes show ORN class-specic axon sorting. Dscam mutant ORN47a axons converge in the
neighbourhood of ORN22a glomerulus but the axons never intermingle (D). Neighbouring projecting ORN
classes 47b and 88a show distinct boundaries in wild type (E,E’) as well as in Dscam mutants (F,F’). (G,H)
Dscam mutant axons coming from the maxillary palps converge into distinct ectopic spots at the border to the
AL or at the suboesophagial ganglion (SOG) (asterisks in H”). Ectopic projecting ORN46a axons converge
into one distinct spot in the neighborhood of other ectopic spots (H’) and they never intermingle with other
ORN classes e.g. ORN71a (J). SOG: suboesophagial ganglion. Green: (A–J) sytGFP; blue: (A,B) N-Cad, (C–J)
Toto3; red: (A–J) ratCD2. Scale bar: 25 µm. Genotype: (A) eyp UAS-CD2; FRT42 47a::sytGFP/FRT42 PCNA;
47a-Gal4 UAS-CD2. (B) eyp UAS-CD2; FRT42 Dscam 47a::sytGFP/FRT42 PCNA; 47a-Gal4 UAS-CD2. (C)
eyp UAS-CD2; FRT42 47a::sytGFP/FRT42 PCNA; 22a-Gal4 UAS-CD2. (D) eyp UAS-CD2; FRT42 Dscam
47a::sytGFP/FRT42 PCNA; 22a-Gal4 UAS-CD2. (E) eyp UAS-CD2; FRT42 47b::sytGFP/FRT42 PCNA; 88a-
Gal4 UAS-CD2. (F) eyp UAS-CD2; FRT42 Dscam 47b::sytGFP/FRT42 PCNA; 88a-Gal4 UAS-CD2. (G) eyp;
FRT42/FRT42 PCNA; 46a::sytGFP/MT14-Gal4 UAS-CD2. (H) eyp; FRT42 Dscam/FRT42 PCNA; 46a::sytGFP/
MT14-Gal4 UAS-CD2 (I) eyp UAS-CD2; FRT42/FRT42 PCNA; 46a::sytGFP 71a-Gal4 UAS-CD2. (J) eyp
UAS-CD2; FRT42 Dscam/FRT42 PCNA; 46a::sytGFP 71a-Gal4 UAS-CD2.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
4
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
On the contrary, removal of Dscam in PNs using MARCM51 (using hs-flp to generate small PN clones,
Supplementary Fig.S5) and targeted Dscam-RNAi expression (Using PN specic GH146-Gal4, Supplementary
Fig.S6) does not aect axonal convergence of wild type ORNs (Supplementary FigsS5and S6; n = 21 for Dscam
MARCM clones; n = 10 for GH146 > DscamRNAi per OR class).
Figure 2. Pre- and postsynaptic recognition is Dscam independent. (A) Schematic showing ORN/PN/LN
matching in olfactory lobes of Drosophila in wild type and Dscam mutants. (B–E) ORN-PN matching identities
remain in Dscam mutants. Mz-19 positive PN dendrites connect to axons of ORN class 88a (D) and not to
47b (B) in wild type. In Dscam mutant, the dendrites follow the misprojecting 88a axons (E) but avoid ectopic
47b axons (C). (F–J) GH146 expressing dendrites do not innervate ectopic Dscam mutant ORN 47a spots,
when they are far away from the wild type glomerulus (box in H, compared to F). Ectopic spots of ORN46a
axons outside of the AL are not innervated by GH146-positive dendrites (G,J) even if the glomerulus of the
ORN46a class is innervated by GH146-positive dendrites. (K–N) Ectopic Dscam mutant spots are innervated
from C753-positive LNs in the AL in case of ORN21a (L) as well as outside the AL in case of ORN46a (N).
Green: sytGFP, red: ratCD2, blue: Toto3. Scale bar: 25 µm. Genotype: (B) eyp UAS-CD2; FRT42/FRT42 PCNA;
47b::sytGFP Mz19-Gal4 UAS-CD2. (C) eyp UAS-CD2; FRT42 Dscam/FRT42 PCNA; 47b::sytGFP Mz19-Gal4
UAS-CD2. (D) eyp UAS-CD2; FRT42/FRT42 PCNA; 88a::sytGFP Mz19-Gal4 UAS-CD2. (E) eyp UAS-CD2;
FRT42 Dscam/FRT42 PCNA; 88a::sytGFP Mz19-Gal4 UAS-CD2. (F–G) eyp UAS-CD2; FRT42 OR::sytGFP/
FRT42 PCNA; GH146-Gal4 UAS-CD2. (H–J) eyp UAS-CD2; FRT42 Dscam OR::sytGFP/FRT42 PCNA; GH146-
Gal4 UAS-CD2. (K,M) eyp UAS-CD2; FRT42 OR::sytGFP/FRT42 PCNA; C753-Gal4 UAS-CD2. (L,N) eyp
UAS-CD2; FRT42 Dscam OR::sytGFP/FRT42 PCNA; C753-Gal4 UAS-CD2.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
5
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
To test if ectopic glomeruli are accessible for postsynaptic neurites we analyzed their interaction with local
interneurons (LNs). In the wild type AL, LNsdo not display glomerulus-specicity but elaborate their neurites
throughout the AL (Fig.2K,M). In ORN-specic large Dscam mosaic brains, each of the ectopic glomeruli inside
the AL receives postsynaptic innervation from LNs (n = 20 per OR class), indicating that the distant ectopic ORN
axons are able to interact with theneurites. Furthermore, even the Dscam mutant ectopic glomeruli that are
formed adjacent to the AL attracts LNneurites (Fig.2K–N; n = 8).
In summary, Dscam mutant glomeruli localized close to the target area receive the correct postsynaptic inner-
vation. During wild type ORN convergence, PN dendrites occupy broad AL region and restrict subsequently onto
a single glomerulus, suggesting that locally displaced protoglomeruli in Dscam mosaics are in contact with the
dendrites of their PN partners13. In contrast, distant ectopic glomeruli do not associate with PN dendrites indi-
cating a high level of ORN-PN recognition specicity in the Drosophila olfactory system, which is not aected in
Dscam mutant ORNs. On the other hand, LN neurites seem to associate with all protoglomeruli, which have been
formed within their range independent of the ORN class identity. As Dscam does not seem to aect the inter-
action of ORN axons with their target neurites we further characterized its role in the inter-axonal recognition.
Formation of ectopic glomeruli requires inter-axonal recognition. To determine if inter-axonal rec-
ognition is required for the formation of ectopic glomeruli, we analyzed the projections of single Dscam mutant
ORNs surrounded by other Dscam mutant or wild type ORNs. First, we used ybow52,53 to visualize individual
mutant Or47a axons in the background of a large ORN specic Dscam mutant eld (generated using eyeless-p/
PCNA). Visualizing individual Dscam mutant ORN47a axons showed that even small ectopic glomeruli consist
of multiple axons, which indicates inter-axonal recognition as a possible mechanism for ectopic axon conver-
gence (Fig.3A,B). Next we visualized axon targeting of small Dscam mutant clones in a Dscam heterozygous
mutant eld using MARCM. is showed that single Dscam mutant ORNs could target normally to the wild
type site (Fig.3C,D, 46/48 1–3 cell clones show WT targetting). is observation combined with results from
Flybow clones and convergence of mutant maxillary ORNs outside antennal lobe (above) showed that formation
of ectopic glomeruli is a result of interactions between axons of the same OR class.
Dscam functions in a cell autonomous manner. To determine how Dscam modulates self-recognition
of axons from the same OR class we performed a series of mosaic and rescue experiments. First, are wild type
axons in genetic mosaics also attracted to coalesce into Dscam mutant ectopic glomeruli? e selective labeling
of the homozygous wild type axons in “reverse MARCM” clones54 revealed that they bypass the ectopic glo-
meruli and project to their wild type target glomerulus (Fig.4A–D, Supplementary Fig.S7, n(ORN47a) = 12,
n(ORN46a) = 6, n(ORN47b) = 8, n(ORN21a) = 12). This result indicated that Dscam functions in a
cell-autonomous manner to prevent inter-axonal recognition. Interestingly, some of the Dscam mutant axons
also reach the WT target site (Fig.4A,B). Second, we tested if the expression of a single Dscam isoform in mutant
ORNs could rescue the ectopic glomerulus formation. An early developmental expression (using elav-Gal4) of
dierent Dscam isoform (e.g. Dscam1.30.30.2 or Dscam11.31.25.2) in Dscam mutant ORNs (generated using eyless-p)
suppresses the ectopic convergence inside and outside the AL (Fig.4G,H). However, we also observed a dis-
ruption of ORN convergence at the target side and axons bypass their target glomerulus (See also Fig.4R,T).
Interestingly, we observed aggregates of Synaptotagmin::GFP along the axonal lengths (Fig.4G arrows). As a
similar axon targeting defect was observed following the expression of the same Dscam isoform in wild type
ORNs (Fig.4E,F), we conclude that the predominant expression of a single Dscam isoform prevents ORN axon
convergence. ese results suggest that the precise level or isoform diversity of Dscam is necessary to allow cor-
rect olfactory system patterning.
Over-expression of a single Dscam isoform blocks axon convergence. To learn more about
the repulsive activity of Dscam in ORN axon convergence, we performed a series of cell-type specic Dscam
over-expression experiments. First, single Dscam isoform expression in all olfactory sensory neurons (using
SG18.1-Gal4) completely disrupts axon convergence leading to an aglomerular AL target eld (Fig.4I,J). Second,
Dscam over-expression in ORN subgroups, which target to distinct regions within the AL (Connectin-gal4
expresses in ORN classes projecting onto the lateral antennal lobe), prevents only the convergence within the
expression domain whereas adjacent glomeruli are not aected (Fig.4K–N, Supplementary Fig.S8). ird, if we
express Dscam early in development (using E132 or 72OK-Gal4) in single (using hs-p generated clones) or all
neurons of an individual ORN class, we observed a normal projection towards their target region but a failure to
coalesce (Fig.4O–T). ese results suggest that the expression of identical isoforms on neighboring ORNs does
not aect their projection towards the target area but neurons do not coalesce into a single glomerulus.
e fact that wild type ORN axons that are in contact with Dscam over-expressing axons are not aected in
their projection pattern suggests that the Dscam-induced axon repulsion is mediated through homophilic iso-
form recognition between interacting ORN axons. is would predict that Dscam expression in a single ORN
will not interfere with the axon projection as shown before for mushroom body55. However, we observed a similar
axon overshoot phenotype by expressing Dscam in a single ORN of dierent classes (Fig.5A–E) in otherwise WT
background. A more detailed single axon analysis showed that the terminal arborization of wild type axons is
strongly reduced in Dscam expressing ORNs (Fig.5B,C,E). In addition, small collateral extension can be observed
in irregular positions along the axon sha.
us, in agreement with the cell-autonomous role of Dscam observed in loss-of-function studies, these
gain-of-function experiments also support a mechanism of axon convergence in which Dscam does not medi-
ate inter-axonal communication directly but Dscam activity on projecting axons suppresses axonal convergence
among ORNs of the same sensory class.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
6
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Regulation of ORN axon convergence during development. Finally, we determined the devel-
opmental step at which Dscam controls ORN axon recognition and convergence. We generated ORN specic
eyless-p clones and used elav-Gal4 for early developmental expression and labeling of ORNs. In wild type, ORN
axons project along the periphery of the AL (Fig.6A, n > 10) where they coalesce into class-specic protoglomer-
uli before they merge with the dendritic eld (Fig.6B, n > 10). e over-expression of a single Dscam isoform
in ingrowing ORN axons fully prevents their initial coalescence into protoglomeruli (Fig.6C,D, Supplementary
Fig.S9, n > 10). ese axons stay in the nerve ber layer and never interact with the dendritic eld. In contrast,
ORN axons mutant for Dscam coalesce into protoglomeruli as soon as they enter the target area thereby prevent-
ing their projection along the AL surface (Fig.6E, n > 10). e number and size of these protoglomeruli increases
during further AL development (Fig.6F, n > 10) but these accumulations of axon terminals fail to interact with
the dendritic eld. ese results show that Dscam is critical in the initial control of ORN axon convergence. e
Dscam activity has to be tightly regulated to guarantee a spatially xed axon convergence as a reduction in Dscam
leads to premature axon recognition whereas an increase in Dscam activity prevents axons convergence at the
target area.
Figure 3. Formation of ectopic glomeruli requires inter-axonal recognition. (A,B) Multicolour axon labeling
using ybow shows targeting of individual mutant ORN 47a axons in WT (A-A”) and Dscam (B-B”) mutants.
e ectopic glomeruli in Dscam mutants always consisted of multiple axons (marked with arrow heads in
B’ and B”). (C,D) Visualizing targeting of single/few cell OR47a clones in WT (C) and Dscam mutants (D)
showed similar phenotype with OR47a axons not mis-targetting in Dscam mutants. e number of ORNs was
conrmed by visualizing cell bodies in the le and right antenna. Blue: Ncad (A,B), Red: Ncad (C,D). Scale bar:
25 µm. Genotype: (A) eyp; FRT42/FRT42 PCNA; UAS-FB1.1, Or47a-Gal4/hs-mp5 (B) eyp; FRT42 dscam21/
FRT42 PCNA; UAS-FB1.1, Or47a-Gal4/hs-mp5 (C) eyp; FRT42/FRT42 PCNA; UAS-FB1.1, Or47a-Gal4/hs-
mp5 (D) hsp; FRT42 dscam21/FRT42 Gal80; Or47a-Gal4, UAS-mCD8::GFP.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
7
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Figure 4. Dscam acts cell-autonomously. (A–H) Ectopic spots inside (A, arrow heads) as well as outside (B,
arrow head) the AL are innervated only from homozygous Dscam mutant axons, whereas homozygous wild
type axons always reach their wild type glomerulus (C,D, dotted circle). Presence of Dscam mutant axons is
indicated by the presence of ectopic spots outside the AL, near the V glomerulus (asterisks). Over-expression
of one single Dscam-isoform in wild type axons leads to disrupted glomerular pattern and axon termini are
spread over a large area (E,F). Over-expression in Dscam mutant axons shows the same phenotype than the
over-expression in wild type (G,H). Only the early stopping phenotype outside the AL of maxillary ORN axons
can be rescued by over-expression of a single Dscam-isoform (H). (I–T) Broad over-expression of a single
Dscam-isoform in ORNs show an AL wide distribution of axon termini (green in J) and a complete loss of the
glomerular structure (red in J). Dscam over-expression aects only ORN axons, in which it is expressed. On
over-expression in the con-positive ORNs, the structure of the glomerulus innervated from ORN class 47b
is totally disrupted (L) whereas the ORN class 47a, which project neighboring to the con-positive domain, is
unaected (N). Over-expression of a single Dscam isoform in single ORN axons also reveal a misprojecting
phenotype (P,R,T). Green: sytGFP, blue: Toto3, red: (A–J,O,P) N-cad, (K–N,Q–T) ratCD2. Scale bar: 25 µm.
Genotype: (A,B) eyp; FRT42 Dscam/FRT42 Gal80; OR-Gal4 UAS-sytGFP. (C,D) eyp; FRT42 Dscam Gal80/
FRT42; OR-Gal4 UAS-sytGFP. (E,F) eyp elav-Gal4; FRT42 OR::sytGFP/FRT42 Gal80; UAS-Dscam17.2-7. (G,H)
Content courtesy of Springer Nature, terms of use apply. Rights reserved
8
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Discussion
Sensory maps in the olfactory system are characterized by an astonishing level of synaptic specicity. A multi-
tude of sensory neuron classes, broadly distributed in the peripheral epithelium, project into the CNS primary
target region where they segregate into distinct synaptic glomeruli according to their OR class identity. In the
mouse olfactory system, the odorant receptors, expressed on the projecting ORN axons, are thought to provide
unique recognition identities for the ORN class specic axon sorting in the olfactory bulb2–5. However, ORs are
not directly involved in the inter-axonal recognition but determine the expression of a distinct combination of
cell surface receptors and adhesion molecules that provide spatial information and unique axonal specicity6,7.
It has been shown before that Ig-domain adhesion molecule Dscam plays a critical role in olfactory map for-
mation in Drosophila26. Dscam mutant ORNs frequently mistarget and form ectopic glomeruli26. On the other
hand, Dscam is not involved in synaptic recognition between presynaptic ORNs and post-synaptic PNs45. Here
we show that although ORs are not involved in sensory neuron connectivity, ORN axons also carry unique iden-
tity tags for ORN class-specic recognition independent of their target area. Dscam is critical for the precise
regulation of inter-axonal recognition but does not provide recognition identity itself. Loss of Dscam leads to
class-specic convergence before ORN axons reach their target area. In contrast, Dscam over-expression in wild
type suppresses the ORN axon convergence at ectopic sites as well as at the presumptive target area. Overall,
Dscam mediated inhibition of inter-axonal recognition in developing ORNs of same OR class seems to be con-
trolled by a cell intrinsic mechanism (similar to Dscam mediated self-avoidance) instead of inter-axonal binding
of Dscam between ORN axons of same OR class.
From these data we propose a model of how a balanced Dscam activity controls spatially precise ORN conver-
gence (Fig.6G). Dscam expression on projecting ORN axons prevents the premature recognition between axons
of the same OR class which would lead to the coalescence into protoglomeruli. e Dscam activity has to be over-
come as ORN axons enter the target area to allow axon coalescence into OR class-specic protoglomeruli. e
signal is most likely derived from the target eld as PN dendrites are known to build a pre-patterned dendritic
eld before ORN axon arrival18. e predicted spatial cues could function through the down-regulation of Dscam
signaling or enforce inter-axonal adhesion independent of Dscam31,42,43,56–61. Later self-recognition between iden-
tical Dscam isoforms expressed on individual ORs ensures proper glomerular maturation and synaptogenesis.
The role of Dscam diversity in olfactory system development. e broad expression of a single
Dscam isoform in wild type and Dscam mutant ORNs allows a normal projection towards their target area but
disrupts axon convergence at the prospective target side. is indicates that Dscam isoform diversity is impor-
tant to prevent inter-axonal repulsion, which would interfere with the target-induced axon convergence. Here,
inter-axonal Dscam binding and recognition is prevented through the expression of non-overlapping sets of
Dscam isoforms. e phenotype described in our Gal4-driven over-expression studies resembles the olfactory
connectivity defect observed in transgenically engineered ies that express only a single Dscam isoform in all
neurons46. As we observed the same axonal phenotype with Gal4 lines of dierent expression strength, increased
Dscam signaling due to an enhanced homophilic isoform binding is most likely the cause of the induced axon
targeting defects. e lack of early ORN class specic marker lines has prevented the identication of the Dscam
isoform expression prole in distinct sensory neurons classes so far. However, based on earlier expression analy-
sis26 and identication of isoform expression proles from other neuronal cell types37,62 suggests that each ORN
expresses a unique set of Dscam isoforms, which could be more similar among members of one ORN class com-
pared to ORNs of dierent ORN classes. As our results indicate a cell-autonomous function of Dscam, the actual
combination of Dscam isoforms expressed by a single ORN can still rather be stochastic as a coordinated expres-
sion between all members of the same ORN class is not necessary.
Surprisingly, the preferential expression of a single Dscam isoform in individual ORNs induces an overgrowth
phenotype. us, in contrast to the role of Dscam in axon branch segregation37,46,55 or dendritic eld pattern-
ing38,45, the signaling activity of Dscam in ORN axons has to be precisely regulated to allow spatially dened
convergence. Alternatively, the unique isoform combination expressed on each ORN axon provide a balanced
Dscam activity which is sucient to prevent en passant inter-axonal recognition but can be overcome by exter-
nal derived signals once the axon enters the target area. e intra-neuronal Dscam signaling could be induced
through a binding of the same isoform in trans (on adjacent lopodia) or via cis-clustering in a single lopodia.
Recent structural analyses and in vitro studies have shown that homophilic binding requires the assembly of
Dscam isoforms in larger molecular clusters34,63.
Developmental mechanisms in olfactory system formation. Sensory maps in Drosophila and mam-
mals are characterized by a similar structural organization. In the visual system, photoreceptor neurons pro-
jections into the brain visual centers are patterned in a topographic fashion, whereas sensory neurons in the
olfactory system segregate according to their OR identity into a discrete synaptic map14,17. However, for the for-
mation of the visual and olfactory map there seems to be dierent developmental control mechanisms employed
in ies and mammals. In the mouse visual system, sensory neurons are guided predominantly by gradients of a
eyp elav-Gal4; FRT42 Dscam OR::sytGFP/FRT42 Gal80; UAS-Dscam17.2-7. (I) SG18.1-Gal4/OR::sytGFP. (J)
SG18.1-Gal4/OR::sytGFP; UAS-Dscam17.2-7. (K,M) OR::sytGFP; con-Gal4 UAS-CD2. (L,N) OR::sytGFP; con-
Gal4 UAS-CD2/UAS-Dscam17.2-7, (O) hsp E132-Gal4 UAS-sytGFP; FRT42 Gal80/FRT42. (P) hsp E132-Gal4
UAS-sytGFP; FRT42 Gal80/FRT42;UAS-Dscam17.2-7. (Q) OK72-Gal4 UAS-CD2, (R) eyp; FRT42 OK72-Gal4
UAS-CD2/FRT42 Gal80; UAS-Dscam17.2-7, (S) hsp; FRT42 OK72-Gal4 UAS-CD2/FRT42 Gal80. (T) hsp; FRT42
OK72-Gal4 UAS-CD2/FRT42 Gal80; UAS-Dscam17.2-7.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
9
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
few secreted factors to the correct location in the target area64,65 whereas in ies direct axon-target interactions
seems to organize R-cell type specic innervation66–69. In the mouse olfactory system, the odorant receptors,
expressed on projecting sensory axons, provides unique recognition identities via the regulation of cell-cell sign-
aling molecules6. Here we show that olfactory neurons in Drosophila also possess distinct axonal identities inde-
pendent of OR regulation. Similar to the developing mouse olfactory system, self-organization of the sensory
map formation through aerent-aerent interaction seems to be the primary control mechanism. In addition, the
precise ORN axon targeting requires two independent regulatory mechanisms, one which generates class-specic
ORN axon identities (trans cellular signaling) and a second mechanisms which modulates the activity of this
axon recognition (cell autonomous signaling). Finally, a certain degree of stochastic expression can be found in
both olfactory systems, namely the OR receptor choice in mouse and Dscam isoform expression in ies. On the
other hand, in ies, the target region seems to be also involved in the modulation of inter-axonal recognition and
we demonstrate unique axon-target recognition in ORN-PN matching. It will be interesting to determine in the
future if axon-axon and axon-dendrite interactions are using the same surface recognition code and how this is
regulated via Dscam signaling.
Dscam regulates inter-axonal communication. Interactions between targeting axons is gaining more
and more importance as the self-organizing theme underlying olfactory circuit formation70. e most compelling
evidence comes from mice, where ORNs are able to converge and maintain the gross oderotopic map in the back-
ground of an ablated Olfactory bulb10,71. Consistent with this, we also found that regulation of interactions between
targeting axons is essential for proper Oderotopic map formation in Drosophila. In addition to its well-established
cell intrinsic role in self neurite repulsion72, Dscam regulates inter-axonal interactions between ORNs belonging
to same OR class. ough not directly involved, it might cell intrinsically regulate levels of other adhesion mole-
cules which are involved in recognition between ORNs of same OR class. ree observations directly support this
hypothesis, (1) formation of ectopic glomeruli outside antennal lobe in Dscam mutants, not innervated by PNs,
(2) Flybow showing multiple axons in Dscam mutant ectopic glomeruli and (3) single cell Dscam mutant clones
showing no mis-targeting defects. How Dscam achieves regulating inter-axonal interactions is not clear. It might
regulate the expression of signaling/adhesion molecules involved in class specic ORN convergence, which itself
might depend on a complex combinatorial code. Dscam mediated axonal interactions is one aspect of the multiple
processes involved in olfactory circuit formation and will integrate with aspects like Sema/Plexins30, Epherins73,
Figure 5. Over-expression of single Dscam isoform prevents targeting in a single neuron. (A,D) Single
labelled wild type ORN axons from the antenna (A) and the maxillary palp (D) showing branching inside one
glomerulus (A’,D ’) and the contralateral branch. (B,C,E) Over-expression of a single Dscam isoform leads to
defective branching inside the WT glomerulus (B’,E ’ ) as well as Dscam mutant (C’). White: mCD8::GFP, Red:
N-cad. Scale bar: 25 µm. Genotype: (A,D) hsp, elav-Gal4 UAS-mCD8::GFP; FRT42 Gal80/FRT42. (B,E) hsp,
elav-Gal4 UAS-mCD8::GFP; FRT42 Gal80/FRT42; UAS-Dscam17.2-7. (C) hsp, elav-Gal4 UAS-mCD8::GFP;
FRT42 Gal80/FRT42 Dscam; UAS-Dscam17.2-7.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
10
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Teneurins74 or DIPs/DPRs75 mediated specication of olfactory connectivity or Glia mediated sorting of ORNs as
has been shown in Manduca76. Interestingly in Dscam mutant clones, some mutant ORNs always reach wild type
site. is can result from some axons escaping inter-axonal interactions, thus reaching wild type site. Some other
possibilities can be a redundancy in the function of various Dscam isoforms26 or activity of some recognition mol-
ecules in a complex combinatorial code for ORNs targeting may allow some ORNs to target to their wild type sites.
Methods
Genetics. Fly stocks were maintained in standard medium at 25 °C unless stated otherwise. ree dierent
dscam alleles, reported as null alleles, were used for the analysis: dscam21, dscam23, dscam3326. dscam21 allele was
used in all experiments in addition to either dscam23 or dscam33.
Figure 6. Dscam function is needed in early pupal development. At 25–30 h APF (aer puparium formation)
the ORN axons surround the developing AL before growing into the center(A). Over-expression of a single
Dscam isoform leads to a compaction of the surrounding axon tract (C) whereas in Dscam mutants the ectopic
pre-stopping and ectopic convergence is already visible (E). During 45–50 h APF in wild type the axons grow
into the central part of the AL and start the glomerulus formation (B), over-expression of Dscam prevents the
growth of the axons into the central part of the AL (D). In Dscam mutants the amount of ectopic spots increases
(F). Green: mCD8::GFP; red: N-cad. Scale bar: 25 µm. Genotype: (A,B) eyp; FRT42/FRT42 Gal80; elav-Gal4
UAS-mCD8::GFP. (C,D) eyp; FRT42 UAS-Dscam17.2-1/FRT42 Gal80; elav-Gal4 UAS-mCD8::GFP. (E,F) eyp;
FRT42 Dscam/FRT42 Gal80; elav-Gal4 UAS-mCD8::GFP.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
11
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Markers for dierent ORN subclasses and AL neurons. To label ORNs of one single olfactory class we
used the following promoters fused to Gal4: Or22a, Or46a, Or47a14, Or71a26 and Or88a77. e reporters to visual-
ize axons and synaptic terminals were UAS-mCD8-GFP51, UAS-Synaptotagmin-GFP78 and UAS-rCD279. To visual-
ize ORNs in mosaics or/and to over-express a single Dscam isoform the enhancer trap line Gal4-C155 (elav-Gal4,
all neurons80), SG18.1-Gal4 (all ORNs)81, con-Gal4 (lateral ORN classes)29, MT14-Gal4 (olfactory, gustatory and
mechano-sensory neurons)82, E132-Gal4 (restricted expression in eye-antennal disc)83 and OK72-Gal4 (antennal
lobe glomerulus VM1 and VM4)27 were used. e C753-Gal4 lines are expressed in LNs84. Projection neurons
were visualized with the enhancer trap lines GH146-Gal4 and Mz19-Gal485. For simultaneous visualization of
two ORN classes or target neurons and ORN terminals the Synaptotagmin-GFP was directly expressed under Or
promotor control15,27.
Genetic mosaics. All of the genetic mosaics were generated using the FRT/FLP system86 with various Gal4
drivers (see previous section). For large clones in the antenna and maxillary palps, an eyless-FLP insertion on
the X chromosome in combination with cell lethal PCNA (proliferating cell nuclear antigen, an auxillary pro-
tein of DNA polymerase Delta and control eukaryotic DNA replication) on Dscam+ chromosome was used47,48.
Eyless-p will cause recombination in half of the ORNs generating Dscam−/− ORNs, but not in the central brain
cells leaving all other CNS neurons Dscam−/+. For small clones and single-cell analysis, an hsp70-FLP transgene
on the X chromosome was used87. To visualize the homozygous mutant ORNs, the MARCM system51 with vari-
ous Gal4 drivers (see previous section) and FRT42 TubP-Gal80 was used. To visualize ORN subclasses, mosaics
were generated in ies of the following genotype: eyFLP; FRT42/FRT42 Gal80; Or-Gal4 UAS-sytGFP and eyFLP;
FRT42 dscam/FRT42 Gal80; Or-Gal4 UAS-sytGFP. “Reverse MARCM” genotype was as follows: eyFLP; FRT42
dscam Gal80/FRT42; Or-Gal4 UAS-sytGFP. To visualize ORN terminals and target cell dendrites, we used the
following genotype: eyFLP; Or::sytGFP FRT42/FRT42 PCNA; C753-Gal4 UAS-CD2 and eyFLP; FRT42 dscam
Or::sytGFP/FRT42 PCNA; C753-Gal4 UAS-CD2, eyFLP; GH146-Gal4FRT42/FRT42 PCNA; Or::sytGFP UAS-CD2,
eyFLP; FRT42 dscam GH146-Gal4/FRT42 PCNA; Or::sytGFP UAS-CD2. Double labelling of two ORN single
classes one Or::sytGFP fusion construct and one Or-Gal4 UAS-CD2 line is used as follows: eyFLP UAS-CD2;
FRT42 Or::sytGFP/FRT42 PCNA; Or-Gal4 UAS-CD2 and eyFLP UAS-CD2; FRT42 dscam Or::sytGFP/FRT42
PCNA; Or-Gal4 UAS-CD2. To label the projection of single neurons in wild type and with over-expression of
a single dscam isoform, this genotype was used: hsFLP, elav-Gal4, UAS-mCD8GFP; FRT42 Gal80/FRT42 and
hsFLP, elav-Gal4, UAS-mCD8GFP; FRT42 Gal80/FRT42 UAS-dscam17.2-7. Single-cell clones were obtained by heat
shocking late third-instar larvae (10 min at 37 °C). Developmental studies with the marker elav-Gal4 were per-
formed using pupae of the following genotypes: eyFLP; FRT42/FRT42 Gal80; elav-Gal4 UAS-CD8GFP and eyFLP;
dscam FRT42/FRT42 Gal80; elav-Gal4 UAS-CD8GFP and eyFLP elav-Gal4; FRT42 Gal80/FRT42 UAS-dscam17.2-7.
For all these experiments, we used Dscam21 and Dscam33 in parallel26.
Immunohistology. Primary antibodies used for immunohistochemistry were: rat anti-N-Cadherin extracel-
lular domain (DN-Ex #8; 1:2088, 1997, DSHB); rabbit anti-GFP (1:1000; Molecular Probes); and mouse anti-CD2
(1:1000; Molecular Probes). Secondary antibodies used were as follows (all 1:300): goat anti-rabbit F(ab)′ frag-
ment coupled to Alexa 488 (Molecular Probes), goat anti-mouse F(ab)′ fragment coupled to Alexa 568 (Molecular
Probes), goat anti-mouse F(ab)′ fragment coupled to Alexa 568 highly cross-absorbed (Molecular Probes), goat
anti-rat F(ab)′ fragment coupled to Alexa 568 (Molecular Probes), goat anti-rat F(ab)′ fragment coupled to Alexa
647 (Molecular Probes) and Toto-3 (1:2000, Molecular Probes). Immunostaining of brains of adult ies and
pupae were carried out essentially as described previously89 with the following exceptions: (1) adult brains were
xed in 2% PFA for 90 min, and (2) for the dissection of the pupal brains, the pupal cases were open, 2% PFA
was added, and the brains were allowed to x for 10 min before further dissection in 2% PFA. e overall time of
xation in 2% PFA was 90 min. Fluorescent samples were analyzed using a Zeiss Meta510 confocal microscope.
Image processing. e majority of the images were processed using Fiji®90. For single cell clones, the stacks
of confocal images were rst 3D rendered in Imaris®. en with the “Surface” tool, the innervation site of the
axon in antennal lobe and the cell bodies in antenna were reconstructed by thresholding intensity. en the
axonal bers of the neuron were reconstructed using the tool “Filaments”. is was carried out manually because
of the low intensity of GFP in the axons and the high background noise.
Data Availability
All the data generated and analyzed during this study are included in this published article and its supplementary
information les.
References
1. omiyama, T. & Luo, L. Development of wiring specicity in the olfactory system. Curr. Opin. Neurobiol. 16, 67–73 (2006).
2. Mombaerts, P. et al. Visualizing an olfactory sensory map. Cell 87, 675–686 (1996).
3. Wang, F., Nemes, A., Mendelsohn, M. & Axel, . Odorant receptors govern the formation of a precise topographic map. Cell 93,
47–60 (1998).
4. Feinstein, P., Bozza, T., odriguez, I., Vassalli, A. & Mombaerts, P. Axon guidance of mouse olfactory sensory neurons by odorant
receptors and the B2A. Cell 117, 833–846 (2004).
5. Feinstein, P. & Mombaerts, P. A contextual model for axonal sorting into glomeruli in the mouse olfactory system. Cell 117, 817–831
(2004).
6. Imai, T., Suzui, M. & Saano, H. Odorant receptor-derived cAMP signals direct axonal targeting. Science 314, 657–61 (2006).
7. Serizawa, S. et al. A Neuronal Identity Code for the Odorant eceptor-Specic and Activity-Dependent Axon Sorting. Cell 127,
1057–1069 (2006).
8. Bulfone, A. et al. An olfactory sensory map develops in the absense of normal projection neurons or GABAergic interneurons.
Neuron 21, 1273–1282 (1998).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
12
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
9. Graziadei, P. P., Levine, . . & Graziadei, G. A. egeneration of olfactory axons and synapse formation in the forebrain aer
bulbectomy in neonatal-mice. Proc. Natl. Acad. Sci. USA 75, 5230–5234 (1978).
10. St. John, J. A., Clarris, H. J., Mceown, S., oyal, S. & ey, B. Sorting and convergence of primary olfactory axons are independent
of the olfactory bulb. J. Comp. Neurol. 464, 131–140 (2003).
11. Mombaerts, P. Axonal Wiring in the Mouse Olfactor y System. Annu. ev. Cell Dev. Biol. 22, 713–737 (2006).
12. Imai, T. & Saano, H. oles of odorant receptors in projecting axons in the mouse olfactory system. Curr. Opin. Neurobiol. 17,
507–515 (2007).
13. Jeeris, G. S. X. E. & Hummel, T. Wiring specicity in the olfactory system. Semin. Cell Dev. Biol. 17, 50–65 (2006).
14. Vosshall, L. B., Wong, A. M. & Axel, . An olfactory sensory map in the y brain. Cell 102, 147–59 (2000).
15. Fishilevich, E. & Vosshall, L. B. Genetic and functional subdivision of the Drosophila antennal lobe. Curr. Biol. 15, 1548–1553
(2005).
16. Couto, A., Alenius, M. & Dicson, B. J. Molecular, anatomical, and functional organization of the Drosophila olfactory system. Cur r.
Biol. 15, 1535–47 (2005).
17. Gao, Q., Yuan, B. & Chess, A. Convergent projections of Drosophila olfactory neurons to specic glomeruli in the antennal lobe.
Nat. Neurosci. 3, 780–5 (2000).
18. Jeeris, G. S. X. E., Marin, E. C., Stocer, . F. & Luo, L. Target neuron prespecication in the olfactory map of Drosophila. Nature
414, 204–208 (2001).
19. Wilson, . I. ole of GABAergic Inhibition in Shaping Odor-Evoed Spatiotemporal Patterns in the Drosophila Antennal Lobe. J.
Neurosci. 25, 9069–9079 (2005).
20. Stocer, . F. e organization of the chemosensory system in Drosophila melanogaster: a rewiew. Cell and Tissue esearch 275,
3–26 (1994).
21. Dobritsa, A. A., van der Goes van Naters, W., Warr, C. G., Steinbrecht, . A. & Carlson, J. . Integrating the molecular and cellular
basis of odor coding in the Drosophila antenna. Neuron 37, 827–41 (2003).
22. ay, A., van der, G., van Naters, W., Shiraiwa, T. & Carlson, J. . Mechanisms of Odor eceptor Gene Choice in Drosophila. Neuron
53, 353–369 (2007).
23. Goldman, A. L., Van Der Goes Van Naters, W., Lessing, D., Warr, C. G. & Carlson, J. . Coexpression of two functional odor
receptors in one neuron. Neuron 45, 661–666 (2005).
24. Elmore, T., Ignell, ., Carlson, J. . & Smith, D. P. Targeted mutation of a Drosophila odor receptor denes receptor requirement in
a novel class of sensillum. J. Neurosci. 23, 9906–12 (2003).
25. Hong, W. & Luo, L. Genetic control of wiring specicity in the y olfactory system. Genetics 196, 17–29 (2014).
26. Hummel, T. et al. Axonal targeting of olfactory receptor neurons in Drosophila is controlled by Dscam. Neuron 37, 221–31 (2003).
27. Hummel, T. & Zipursy, S. L. Aerent induction of olfactory glomeruli requires N-cadherin. Neuron 42, 77–88 (2004).
28. Jhaveri, D., Saharan, S., Sen, A. & odrigues, V. Positioning sensory terminals in the olfactory lobe of Drosophila by obo signaling.
Development 131, 1903–12 (2004).
29. Lattemann, M. et al. Semaphorin-1a controls receptor neuron-specic axonal convergence in the primary olfactory center of
Drosophila. Neuron 53, 169–84 (2007).
30. Sweeney, L. B. et al. Temporal Target estriction of Olfactory eceptor Neurons by Semaphorin-1a/PlexinA-Mediated Axon-Axon
Interactions. Neuron 53, 185–200 (2007).
31. Schmucer, D. et al. Drosophila Dscam is an axon guidance receptor exhibiting extraordinary molecular diversity. Cell 101, 671–84
(2000).
32. Zipursy, S. L., Wojtowicz, W. M. & Hattori, D. Got diversity? Wiring the y brain with Dscam. Trends in Biochemical Sciences 31,
581–588 (2006).
33. Wojtowicz, W. M., Flanagan, J. J., Millard, S. S., Zipursy, S. L. & Clemens, J. C. Alternative splicing of Drosophila Dscam generates
axon guidance receptors that exhibit isoform-specic homophilic binding. Cell 118, 619–33 (2004).
34. Wojtowicz, W. M. et al. A vast repertoire of Dscam binding specicities arises from modular interactions of variable Ig domains. Cell
130, 1134–1145 (2007).
35. Wu, W., Ahlsen, G., Baer, D., Shapiro, L. & Zipursy, S. L. Complementary chimeric isoforms reveal Dscam1 binding specicity in
vivo. Neuron 74, 261–8 (2012).
36. Matthews, B. J. et al . Dendrite self-avoidance is controlled by Dscam. Cell 129, 593–604 (2007).
37. Zhan, X.-L. L. et al. Analysis of Dscam diversity in regulating axon guidance in Drosophila mushroom bodies. Neuron 43, 673–86
(2004).
38. Shi, L., Yu, H.-H., Yang, J. S. & Lee, T. Specic Drosophila Dscam juxtamembrane variants control dendritic elaboration and axonal
arborization. J. Neurosci. 27, 6723–8 (2007).
39. Chen, B. E. et al. e molecular diversity of Dscam is functionally required for neuronal wiring specicity in Drosophila. Cell 125,
607–20 (2006).
40. Soba, P. et al. Drosophila sensory neurons require Dscam for dendritic self-avoidance and proper dendritic eld organization.
Neuron 54, 403–16 (2007).
41. Hughes, M. E. et al. Homophilic Dscam interactions control complex dendrite morphogenesis. Neuron 54, 417–27 (2007).
42. im, J. H., Wang, X., Coolon, . & Ye, B. Dscam expression levels determine presynaptic arbor sizes in Drosophila sensory neurons.
Neuron 78, 827–38 (2013).
43. Cvetovsa, V., Hibbert, A. D., Emran, F. & Chen, B. E. Overexpression of Down syndrome cell adhesion molecule impairs precise
synaptic targeting. Nat. Neurosci. 16, 677–82 (2013).
44. He, H. et al. Cell-intrinsic requirement of Dscam1 isoform diversity for axon collateral formation. Science 344, 1182–6 (2014).
45. Zhu, H. et al. Dendritic patterning by Dscam and synaptic partner matching in the Drosophila antennal lobe. Nat. Neurosci. 9,
349–55 (2006).
46. Hattori, D. et al. Dscam diversity is essential for neuronal wiring and self-recognition. Nature 449, 223–7 (2007).
47. Newsome, T. P., Asling, B. & Dicson, B. J. Analysis of Drosophila photoreceptor axon guidance in eye-specic mosaics. Development
127, 851–60 (2000).
48. Yamaguchi, M. et al. Drosophila proliferating cell nuclear antigen (cyclin) gene: structure, expression during development, and
specic binding of homeodomain proteins to its 5’-aning region. Mol. Cell. Biol. 10, 872–9 (1990).
49. Marin, E. C., Jeeris, G. S. X. E. X. E., omiyama, T., Zhu, H. & Luo, L. epresentation of the glomerular olfactory map in the
Drosophila brain. Cell 109, 243–55 (2002).
50. Berdni, D., Chihara, T., Couto, A. & Luo, L. Wiring stability of the adult Drosophila olfactory circuit aer lesion. J. Neurosci. 26,
3367–76 (2006).
51. Lee, T. & Luo, L. Mosaic analysis with a repressible cell marer for studies of gene function in neuronal morphogenesis. Neuron 22,
451–61 (1999).
52. Shimosao, N., Hadjieconomou, D. & Salecer, I. Flybow to dissect circuit assembly in the Drosophila brain. Methods Mol. Biol.
1082, 57–69 (2014).
53. Hadjieconomou, D. et al. Flybow: genetic multicolor cell labeling for neural circuit analysis in Drosophila melanogaster. Nat.
Methods 8, 260–6 (2011).
Content courtesy of Springer Nature, terms of use apply. Rights reserved
13
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
54. Zhu, H. & Luo, L. Diverse functions of N-cadherin in dendritic and axonal terminal arborization of olfactory projection neurons.
Neuron 42, 63–75 (2004).
55. Wang, J. et al. Transmembrane/juxtamembrane domain-dependent Dscam distribution and function during mushroom body
neuronal morphogenesis. Neuron 43, 663–72 (2004).
56. Andrews, G. L. et al. Dscam guides embryonic axons by Netrin-dependent and -independent functions. Development 135,
3839–3848 (2008).
57. Liu, G. et al. DSCAM functions as a netrin receptor in commissural axon pathnding. Proc. Natl. Acad. Sci. USA 106, 2951–6 (2009).
58. Matthews, B. J. & Grueber, W. B. Dscam1-mediated self-avoidance counters netrin-dependent targeting of dendrites in Drosophila.
Curr. Biol. 21, 1480–7 (2011).
59. Worby, C. A. et al. e sorting nexin, DSH3PX1, connects the axonal guidance receptor, Dscam, to the actin cytoseleton. J. Biol.
Chem. 276, 41782–9 (2001).
60. Li, W. & Guan, .-L. e Down syndrome cell adhesion molecule (DSCAM) interacts with and activates Pa. J. Biol. Chem. 279,
32824–31 (2004).
61. Oumura, M., Sauma, C., Miura, M. & Chihara, T. Lining cell surface receptors to microtubules: tubulin folding cofactor D
mediates Dscam functions during neuronal morphogenesis. J. Neurosci. 35, 1979–90 (2015).
62. Neves, G., Zucer, J., Daly, M. & Chess, A. Stochastic yet biased expression of multiple Dscam splice variants by individual cells. Nat.
Genet. 36, 240–6 (2004).
63. Shapiro, L. Self-recognition at the atomic level: understanding the astonishing molecular diversity of homophilic Dscams. Neuron
56, 10–3 (2007).
64. Brown, A. et al. Topographic mapping from the retina to the midbrain is controlled by relative but not absolute levels of EphA
receptor signaling. Cell 102, 77–88 (2000).
65. Monschau, B. et al. Shared and distinct functions of AGS and ELF-1 in guiding retinal axons. EMBO J. 16, 1258–1267 (1997).
66. Choe, .-M., Praash, S., Bright, A. & Clandinin, T. . Liprin-alpha is required for photoreceptor target selection in Drosophila.
Proc. Natl. Acad. Sci. USA 103, 11601–11606 (2006).
67. Clandinin, T. . et al. Drosophila LA regulates 1-6 and 7 target specicity in the visual system. Neuron 32, 237–248 (2001).
68. Lee, C. H., Herman, T., Clandinin, T. ., Lee, . & Zipursy, S. L. N-cadherin regulates target specicity in the Drosophila visual
system. Neuron 30, 437–50 (2001).
69. Lee, . C. et al. e protocadherin Flamingo is required for axon target selection in the Drosophila visual system. Nat. Neurosci. 6,
557–563 (2003).
70. Imai, T. & Saano, H. Axon-axon interactions in neuronal circuit assembly: lessons from olfactory map formation. Eur. J. Neurosci.
34, 1647–54 (2011).
71. Imai, T. et al. Pre-target axon sorting establishes the neural map topography. Science 325, 585–90 (2009).
72. Zipursy, S. L. & Grueber, W. B. e molecular basis of self-avoidance. Annu. ev. Neurosci. 36, 547–68 (2013).
73. aneo, M. & Nighorn, A. Interaxonal Eph-ephrin signaling may mediate sorting of olfactory sensory axons in Manduca sexta. J.
Neurosci. 23, 11523–38 (2003).
74. Hong, W., Mosca, T. J. & Luo, L. Teneurins instruct synaptic partner matching in an olfactory map. Nature 484, 201–7 (2012).
75. Barish, S. et al. Combinations of DIPs and Dprs control organization of olfactory receptor neuron terminals in Drosophila. PLoS
Genet. 14, e1007560 (2018).
76. Gibson, N. J., Tolbert, L. P. & Oland, L. A. Activation of glial FGFs is essential in glial migration, proliferation, and survival and in
glia-neuron signaling during olfactory system development. PLoS One 7, e33828 (2012).
77. omiyama, T., Carlson, J. . & Luo, L. Olfactory receptor neuron axon targeting: Intrinsic transcriptional control and hierarchical
interactions. Nat. Neurosci. 7, 819–825 (2004).
78. Zhang, Y. Q., odesch, C. . & Broadie, . Living synaptic vesicle marer: synaptotagmin-GFP. Genesis 34, 142–145 (2002).
79. Dunin-Borowsi, O. M. & Brown, N. H. Mammalian CD2 is an eective heterologous marer of the cell surface in Drosophila. De v.
Biol. 168, 689–93 (1995).
80. Lin, D. M. & Goodman, C. S. Ectopic and increased expression of Fasciclin II alters motoneuron growth cone guidance. Neuron 13,
507–523 (1994).
81. Jhaveri, D., Sen, A. & odrigues, V. Mechanisms underlying olfactory neuronal connectivity in Drosophila-the atonal lineage
organizes the periphery while sensory neurons and glia pattern the olfactory lobe. Dev. Biol. 226, 73–87 (2000).
82. Tissot, M., Gendre, N., Hawen, A., Störtuhl, . F. & Stocer, . F. Larval chemosensory projections and invasion of adult aerents
in the antennal lobe of Drosophila. J. Neurobiol. 32, 281–97 (1997).
83. Halder, G., Callaerts, P. & Gehring, W. J. Induction of ectopic eyes by targeted expression of the eyeless gene in Drosophila. Science
267, 1788–92 (1995).
84. Yang, M. Y., Armstrong, J. D., Vilinsy, I., Strausfeld, N. J. & aiser, . Subdivision of the Drosophila mushroom bodies by enhancer-
trap expression patterns. Neuron 15, 45–54 (1995).
85. Jeeris, G. S. X. E. et al. Developmental origin of wiring specicity in the olfactory system of Drosophila. Development 131, 117–30
(2004).
86. Xu, T. & ubin, G. M. Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 117, 1223–1237 (1993).
87. Golic, . G. & Lindquist, S. e FLP recombinase of yeast catalyzes site-specic recombination in the drosophila genome. Cell 59,
499–509 (1989).
88. Iwai, Y. e t al. Axon patterning requires DN-cadherin, a novel neuronal adhesion receptor, in the drosophila embryonic CNS. Neuron
19, 77–89 (1997).
89. Van Vactor, D. L., Cagan, . L., rämer, H. & Zipursy, S. L. Induction in the developing compound eye of Drosophila: Multiple
mechanisms restrict 7 induction to a single retinal precursor cell. Cell 67, 1145–1155 (1991).
90. Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–82 (2012).
Acknowledgements
We would like to thank Bloomington stock collection, Prof. Iris Salecker and Prof. Berry Dickson for the y
stocks, Hannah Greshake for experimental help. is work was supported by DFG (Hu992/2-1, SFB 629-B4),
Schram Foundation, Research Platform RoL (University of Vienna) and intramural funds from the University
of Vienna.
Author Contributions
G.G. and T.H. conceptualized the work and wrote the manuscript, G.G., A.Z., M.L., B.B., D.J. and R.K. performed
the experiments and interpreted the results.
Content courtesy of Springer Nature, terms of use apply. Rights reserved
14
SCIENTIFIC REPORTS | (2019) 9:11554 | https://doi.org/10.1038/s41598-019-47924-9
www.nature.com/scientificreports
www.nature.com/scientificreports/
Additional Information
Supplementary information accompanies this paper at https://doi.org/10.1038/s41598-019-47924-9.
Competing Interests: e authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2019
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
Available via license: CC BY 4.0
Content may be subject to copyright.
