doi:10.1038/nature07062
LETTERS
Switch of rhodopsin expression in terminally
differentiated Drosophila sensory neurons
Simon G. Sprecher1 & Claude Desplan1
Specificity of sensory neurons requires restricted expression of
one sensory receptor gene and the exclusion of all others within
a given cell. In the Drosophila retina, functional identity of photoreceptors depends on light-sensitive Rhodopsins (Rhs). The much
simpler larval eye (Bolwig organ) is composed of about 12 photoreceptors, eight of which are green-sensitive (Rh6) and four bluesensitive (Rh5)1. The larval eye becomes the adult extraretinal
‘eyelet’ composed of four green-sensitive (Rh6) photoreceptors2,3.
Here we show that, during metamorphosis, all Rh6 photoreceptors
die, whereas the Rh5 photoreceptors switch fate by turning off Rh5
and then turning on Rh6 expression. This switch occurs without
apparent changes in the programme of transcription factors that
specify larval photoreceptor subtypes. We also show that the transcription factor Senseless (Sens) mediates the very different cellular behaviours of Rh5 and Rh6 photoreceptors. Sens is restricted
to Rh5 photoreceptors and must be excluded from Rh6 photoreceptors to allow them to die at metamorphosis. Finally, we show
that Ecdysone receptor (EcR) functions autonomously both for the
death of larval Rh6 photoreceptors and for the sensory switch of
Rh5 photoreceptors to express Rh6. This fate switch of functioning, terminally differentiated neurons provides a novel, unexpected example of hard-wired sensory plasticity.
The adult Drosophila eyelet comprises approximately four photoreceptors located between the retina and the optic ganglia2. It directly
contacts the pacemaker neurons of the adult fly, the lateral neurons4.
In conjunction with the compound eye and the clock-neuron
intrinsic blue-sensitive receptor cryptochrome3 it helps shift the
phase of the molecular clock in response to light. All eyelet photoreceptors express green-sensitive Rh6, and are derived from photoreceptors of the larval eye2,5,6 that mediate light avoidance and
entrainment of the molecular clock by innervating the larval lateral
neurons7–9.
Larval photoreceptors develop in a two-step process during
embryogenesis1,10. Primary precursors are specified first and develop
as the four Rh5-subtype photoreceptors. They signal through
Epidermal growth factor receptor (EGFR) to the surrounding tissue
to develop as secondary precursors, which develop into the eight
Rh6-subtype photoreceptors1. Two transcription factors specify larval photoreceptor subtypes1. Spalt (Sal) is exclusively expressed in
Rh5 photoreceptors, where it is required for Rh5 expression. Sevenup (Svp) is restricted to Rh6 photoreceptors, where it represses sal
and promotes Rh6 expression. A third transcription factor,
Orthodenticle (Otd), expressed in all larval photoreceptors, acts only
in the Rh5 subtype to promote Rh5 expression and to repress Rh6
(refs 1, 11).
To address the relation between the larval Rh5 and Rh6 photoreceptors and the adult eyelet, we tracked them through metamorphosis (Fig. 1c–e). To permanently label them, we used
UAS-Histone2B::YFP, which is stably incorporated in the chromatin,
1
and thus remains detectable in post-mitotic neurons throughout
pupation12. Surprisingly, all Rh6 photoreceptors degenerate and disappear during early phases of metamorphosis (Fig. 1d). In contrast,
Rh5 photoreceptors can be followed throughout pupation (Fig. 1g–i).
Expression of Rh5 ceases during early stages of pupation and, at midpupation, neither Rh5 nor Rh6 can be detected (Fig. 1h). About four
cells are still present, however, and can be identified by rh5-Gal4/
UAS-H2B::YFP (Fig. 1h) or GMR-Gal4/UAS-H2B::YFP (data not
shown). Eyelet photoreceptors only express Rh6, even though
H2B::YFP driven by rh5-Gal4 is detectable in those cells (Fig. 1i).
Therefore, the four larval Rh5 photoreceptors must switch rhodopsin
expression at metamorphosis to give rise to the four eyelet Rh6
photoreceptors (Fig. 1j). The remaining eight Rh6 photoreceptors
die (Fig. 1f), their axon becoming fragmented before disappearing
(Fig. 1m, n). A ‘memory experiment’ (rh5-Gal4/UAS-Flp;ActFRT . STOP . FRT-nlacZ) also showed that eyelet Rh6 photoreceptors did express Rh5 earlier (Fig. 1k, l).
We further verified the death of Rh6 photoreceptors and transformation of Rh5 photoreceptors by three independent sets of
experiments.
First, we ablated Rh5 photoreceptors by expressing pro-apoptotic
genes rpr and hid (rh5-Gal4/UAS-rpr,UAS-hid). This results in the
absence of larval Rh5 photoreceptors and the complete absence of the
eyelet (Fig. 2g, h)4. Conversely, preventing cell death of the Rh6
subtype by expressing the apoptosis inhibitor p35 (rh6-Gal4/UASp35) leads to an eyelet that consists of 12 photoreceptors, all expressing Rh6 (Fig. 2i).
Second, we blocked development of larval Rh6 photoreceptors by
expressing a dominant negative form of EGFR (so-Gal4/UASH2B::YFP; UAS-EGFRDN) (Fig. 2a)1. The eyelet of these animals is
not affected and three or four cells express Rh6 normally (Fig. 2d).
This shows that larval Rh6 photoreceptors do not contribute to the
eyelet.
Third, we analysed the expression of Sal (Rh5-subtype specific)
and Svp (Rh6-subtype specific) in the adult eyelet: eyelet photoreceptors still express Sal, but not Svp even though these photoreceptors now express Rh6 (Fig. 2b, c, e, f). Rh5 requires Sal expression in
the Bolwig organ, but Otd function is also necessary to activate Rh5
and to repress Rh6. In otd mutants, larval Rh5 photoreceptors
marked by Sal express Rh6 and lack Rh5 expression, thus mimicking
the switch at metamorphosis1. Thus, Rh6 could be expressed in Rh5
photoreceptors if otd function were lost in the eyelet. However, Otd
expression does not change during the transition from the Bolwig
organ to eyelet (data not shown) although it might be inactive in the
eyelet.
What is the trigger that controls the switch from rh5 to rh6?
Ecdysone controls many developmental processes during metamorphosis. EcR is expressed during the third larval instar and pupation in all larval photoreceptors and surrounding tissues (Fig. 3a, b, d,
Center for Developmental Genetics, Department of Biology, New York University, 1090 Silver Center, 100 Washington Square East, New York, New York 10003-6688, USA.
1
©2008 Macmillan Publishers Limited. All rights reserved
LETTERS
NATURE
e)13. To evaluate EcR activity, we used a reporter line in which lacZ is
under the control of multimerized ecdysone response elements
(73EcRE-lacZ)14. The expression of lacZ is absent until late third
instar and prepupation, whereas thereafter all larval photoreceptors
(and surrounding tissue) express 73EcRE-lacZ (Fig. 3c, f). EcR
expression decreases during late pupation and is no longer detectable
by the time Rh6 expression starts in the eyelet (Supplementary Figure
2a, b).
Chp
Rh6
Elav
a
Rh5
Elav
b
rPR
La
Me
Rh6
Rh5
c
YFP
Rh6
GFP
Third instar d
Rh6
Elav
Rh5
Prepupation e
GMR>nYFP
rh5>GFP
YFP
Adult
GMR>nYFP
f
Rh6
g
Rh5
Third instar h
YFP
i
Mid-pupation
rh5>nYFP
rh5>nYFP
Adult
rh5>nYFP
j
Rh6
k
β-Gal
Rh5
Adult
l
rh5>LacZ
Rh6
m
rh5>Flp; Act-STOP-nLacZ
GFP
Elav
Prepupation
rh5>GFP
Adult
Rh6
n
Elav
Prepupation
To test the role of ecdysone, we expressed a dominant negative
form of EcR specifically in larval Rh5 photoreceptors, while permanently labelling these cells (rh5-Gal4/UAS-H2B::YFP;UAS-EcRDN).
This causes no disruption of larval photoreceptor fate, but the eyelet
of these animals now consists of four photoreceptors that all express
Rh5 instead of Rh6 (Fig. 4B, G). A comparable phenotype is observed
after expression of an RNA interference (RNAi) construct for EcR
(rh5-Gal4/UAS-H2B::YFP;UAS-EcRRNAi)(Fig. 4C, G). Therefore,
loss of EcR function prevents larval photoreceptors from switching
to Rh6 expression. In both cases, larval Rh6 photoreceptors still
degenerate and are not observed in the eyelet (Fig. 4G).
We also expressed the dominant negative form of EcR in Rh6
photoreceptors (rh6-Gal4/UAS-H2B::YFP; UAS-EcRDN). In this case,
the Bolwig organ is not affected but the resulting adult eyelet consists
of about 12 photoreceptors, all expressing Rh6 (Fig. 4D, H). This
presumably results from Rh6 photoreceptors not undergoing apoptosis whereas larval Rh5 photoreceptors still switch expression to Rh6
in the eyelet (Fig. 4H). Expression of UAS-EcR-RNAi in Rh6 photoreceptors (rh6-Gal4/UAS-H2B::YFP;UAS-EcRRNAi) leads to the same
results (Fig. 4E, H).
Although EcR could directly control the switch of rhodopsin
expression through binding to the promoters of rh5 and rh6, these
promoters contain no potential EcR binding sites15. Moreover, as no
EcR expression is detectable when Rh6 starts to be expressed, this
would make it unlikely for EcR to control directly the switch to Rh6
(Supplementary Fig. 2a, b). Finally, only allowing expression of the
dominant negative form of EcR starting at mid-pupation (GMRGal4/Tub-Gal80ts,UAS-EcRDN), after rh5 is switched off, does not
prevent activation of Rh6 in the eyelet (Supplementary Fig. 2c, d).
Thus EcR most likely acts in an indirect manner in regulating rhodopsins, likely through the activation of transcription factors that
bind to rh5 and rh6 promoters.
The differential response to ecdysone of Rh6 photoreceptors
(which die) and of Rh5 photoreceptors (which switch to Rh6) must
be due to intrinsic differences between the two subtypes before EcR
signalling. Likely candidates are Sal and Svp. However, late misexpression of Svp in Rh5 photoreceptors (rh5-Gal4/UASH2B::YFP;UAS-svp) or of Sal in Rh6 photoreceptors (rh6-Gal4/
UAS-H2B::YFP;UAS-sal) neither affects rhodopsin expression or cell
number in the eyelet nor alter the expression of rhodopsins in the
Bolwig organ (which is only affected by very early expression of these
transcription factors, through so-Gal4 (ref. 1)). Thus neither Sal nor
Svp are sufficient to alter the response of larval photoreceptors to
EcR.
Figure 1 | Transformation of the larval eye into the adult eyelet. a, b, The
eyelet locates between the optic ganglia (anti-Elav, red; La, lamina; Me,
medulla) and retina (anti-Chp, blue; rPR, retinal photoreceptors). b, The
eyelet only expresses Rh6 (green). c, Larval photoreceptors express Rh6
(green) or Rh5 (blue), nuclei in red (GMR . H2B::YFP). d, Rh6
photoreceptors (green) degenerate during prepupation (arrow), Rh5
photoreceptors (rh5 . GFP, blue) remain, nuclei in red (anti-Elav). e, High
magnification of eyelet photoreceptors expressing Rh6 (green) not Rh5
(blue), nuclei in red (GMR . H2B::YFP). f, Transformation of larval
photoreceptors: Rh6 photoreceptors degenerate, Rh5 photoreceptors
remain at prepupation stages but express Rh6 in the eyelet. g–i, Rh5
photoreceptors tracked through metamorphosis using rh5 . H2B::YFP
(red), anti-Rh6 (green) and anti-Rh5 (blue): g, during third-instar larva;
h, mid-pupation (neither Rh5 nor Rh6 detectable); i, eyelet photoreceptors
now expressing Rh6. j, By mid-pupation, Rh6 photoreceptors have
degenerated whereas Rh5 photoreceptors are now empty, they later switch to
express Rh6. k, No rh5 . lacZ expression can be detected in the eyelet.
l, Genetic memory experiment (rh5-Gal4/UAS-Flp;ActFRT . STOP . FRT-nlacZ): lacZ detected in eyelet (anti b-Gal, red; antiRH6, green). m, n, Projections of Rh6 photoreceptors undergo
fragmentation during prepupation (n, arrow, labelled with anti-Rh6),
whereas Rh5 photoreceptor projections remain (m, Rh5 photoreceptor
projections are shown by rh5-GFP; anti-Elav, red).
2
©2008 Macmillan Publishers Limited. All rights reserved
LETTERS
NATURE
An additional factor, independent from svp and sal, must therefore
allow survival of Rh5 photoreceptors, or promote Rh6 photoreceptor
death. We found that the transcription factor Sens is specifically
expressed in larval Rh5 photoreceptors and remains expressed in
all cells in the eyelet (Fig. 2j, k) where it might act to promote cell
survival. To test this, we misexpressed sens in Rh6 photoreceptors
(rh6-Gal4/UAS-H2B::YFP;UAS-sens). This results in an eyelet that
consists of 12 photoreceptors, all expressing Rh6 (Fig. 2l). Thus,
expression of Sens in Rh6 photoreceptors is sufficient to rescue them
from death, without affecting Sal and Svp expression and subtype
specification of larval photoreceptors (data not shown).
Rh5
Rh6
a
Larva
b
Rh6
Adult
Rh6
Larva
Rh6
Chp
Sens
f
Sal
Adult
Rh6
Elav
Adult
Rh6
k
Sens
YFP
Adult
rh6>nYFP>p35
YFP
Adult
Rh5
i
rh5>hid,rpr
YFP
Larva
Rh5
h
rh5>hid,rpr
j
Rh6
Adult
svp>nYFP
Elav
Rh5
Larva
sal>nYFP
YFP
e
so>nYFP>EGFRDN
g
c
svp>nYFP
so>nYFP>EGFRDN
d
YFP
Larva
Rh6
l
Rh5
YFP
Ecdysone hormonal signalling thus acts in two independent ways
during the formation of the adult eyelet. First, it induces the degeneration of the Rh6 subtype, thereby assuring the correct number of
eyelet photoreceptors. This apoptotic death requires the absence of
Sens, whose expression is restricted to Rh5 photoreceptors that survive. Second, ecdysone signalling is also required to trigger the switch
of spectral sensitivity of blue-sensitive (Rh5) larval photoreceptors to
green-sensitive (Rh6) eyelet photoreceptors (Fig. 4I).
Thus terminally differentiated sensory neurons switch specificity
by turning off one Rhodopsin and replacing it with another.
Although examples of such switches in sensory specificity of terminally differentiated, functional, sensory receptors are extremely rare,
this strategy might be more common than currently anticipated. In
the Pacific pink salmon and rainbow trout, newly hatched fish
express an ultraviolet opsin that changes to a blue opsin as the fish
ages16–18. As in flies, this switch might reflect an adaptation of vision
to the changing lifestyle. The maturing salmon, born in shallow
water, later migrates deeper in the ocean where ultraviolet does not
penetrate. The rhodopsin switch in the eyelet may similarly be an
adaptation to the deeper location of the eyelet within the head, as
light with longer wavelengths (detected by Rh6) penetrates deeper
into tissue than light with shorter wavelengths (detected by Rh5).
The eyelet functions with retinal photoreceptors and
Cryptochrome to entrain the molecular clock in response to light.
The larval eye, on the other hand, functions in two distinct processes:
for the entrainment of the clock and for the larva to avoid light7.
Interestingly, the Rh5 subtype appears to support both functions
whereas Rh6 photoreceptors only contribute to clock entrainment
(S.G.S., J. Blau and C.D., unpublished observations). Thus, the
photoreceptor subtype that supports both functions of the larval
eye is the one that is maintained into the adult and becomes the
eyelet. Why are Rh6-sensitive photoreceptors not maintained? As
these photoreceptors are recruited to the larval eye secondarily, the
ancestral Bolwig organ might have had only Rh5 photoreceptors and
had to undergo a switch in specificity. Larval Rh5 photoreceptors
appear to maintain their overall connectivity to the central pacemaker neurons. However, they are also profoundly restructured
and exhibit widely increased connectivity during metamorphosis.
This might be due to the increase in number of their target neurons,
and the switch of Rh might be part of more extensive plasticity during
formation of the eyelet, including increased connectivity and possibly
the innervation of novel target neurons.
Adult
YFP EcR Elav
a
β-Gal Chp Elav
EcR
c
rh6>nYFP
rh6>nYFP>sens
Figure 2 | Larval Rh5 photoreceptors give rise to the eyelet and express
Rh5 photoreceptor markers. a, Inhibition of larval Rh6 photoreceptor
development (so . H2B::YFP,EGFRDN): only the Rh5 subtype is present in
larvae (blue)1, whereas (d) the eyelet remains unaffected (anti-Rh6, green;
anti-Rh5, blue; anti-YFP, red). b, In the larva, Svp (svp . H2B::YFP, red) is
expressed in Rh6 (green) but not Rh5 photoreceptors (blue), whereas (e) the
eyelet does not express Svp (anti-Rh6, green). c, In the larva, Sal
(sal . H2B::YFP, red) is expressed in Rh5 photoreceptors (blue) but not in
Rh6 photoreceptors (green) (f). Sal (red) is expressed in the eyelet (anti-Rh6,
green). g, h, rh5-Gal4,UAS-hid,rpr ablates Rh5 photoreceptors (g, arrow)
and eyelet (h, arrow) (anti-Rh5, blue; anti Rh6, green; anti Elav, red). i, rh6Gal4/UAS-p35 prevents apoptosis of Rh6 photoreceptors (anti-Rh6, green;
anti-YFP, red). j, k, Sens (red) is detected in the larval Rh5 subtype
(rh5 . H2B::YFP, blue; anti-Chp, green) and the eyelet (rh6 . H2B::YFP;
anti-YFP, blue; anti-Rh6, green). l, Misexpression of UAS-sens in Rh6
photoreceptors (rh6 . H2B::YFP . sens) prevents their apoptosis (antiRh6, green; anti-YFP, red; anti-Rh5, blue). The same phenotype is obtained
with GMR-Gal4;UAS-sens (data not shown).
Third instar
d
Third instar
e
Mid-third instar
f
rh5>nYFP
rh5>nYFP
7xEcRE-lacZ
rh6>nYFP
b
Third instar
Third instar
Late third instar
Figure 3 | EcR expression and activity in larval photoreceptors before
metamorphosis. a, b, d, e, In third-instar larvae, EcR protein (red) is
detected in Rh6 photoreceptors (rh6 . H2B::YFP; green) (a) and in Rh5
photoreceptors (rh5 . H2B::YFP; green) (d); anti-Elav (blue). b, e, The same
as a, d showing only EcR expression (broken circles highlight Rh6
photoreceptors in a and b, Rh5 photoreceptors in d and e). c, f, EcR activity
monitored using 73EcRE-lacZ (b-Gal, green). Anti-Chp (red) and anti-Elav
(blue) marked photoreceptors. No EcR activity is detected during mid-third
instar (c) but is present from late third-instar larval (f) to prepupal stages.
3
©2008 Macmillan Publishers Limited. All rights reserved
LETTERS
NATURE
E
rh5>nYFP
>EcRDN
rh5>nYFP
>EcRRNAi
D
Rh5 YFP
b
Rh6 YFP
F
I
c
Secondary
precursor
Primary
precursor
Embryo
a
rh6>nYFP
>EcRDN
C
a
a
a
rh6>nYFP
>EcRRNAi
B
so>nYFP
Rh6 Rh5 YFP
A
b
c
G
Larva
b
c
Ecdysone
Ecdysone
rh5>EcRDN
sens
a
b
b
c
H
c
Pupa
rhodopsin
switch
rh6>EcRDN
Adult
Figure 4 | EcR is required autonomously for the fate switch of Rh5
photoreceptors and apoptosis of Rh6 photoreceptors. Eyelet expressing soGal4/UAS-H2B::YFP (Aa–Ac), rh5-Gal4,UAS-H2B::YFP, and UAS-EcRDN
(Ba–Bc), rh5-Gal4,UAS-H2B::YFP,UAS-EcRRNAi (Ca–Cc), rh6-Gal4,UASH2B::YFP,UAS-EcRDN (Da–Dc) or rh6-Gal4,UAS-H2B::YFP,UAS-EcRRNAi
(Ea–Ec), anti-Rh6 (green), anti-Rh5 (blue) and anti-YFP (red). Interfering
with EcR in Rh5 photoreceptors prevents cells from switching to Rh6
(b, c). In Rh6 photoreceptors, it prevents apoptosis (D, E). The eyelet (F) after
manipulation of Rh5 (G) or Rh6 photoreceptors (H). I, Transformation of
the larval eye into the eyelet. EcR function leads to apoptosis of the Rh6
photoreceptors and the switch to Rh6 of Rh5 photoreceptors.
The general model that sensory neurons only express a single sensory receptor gene does not hold true for salmon and the fruitfly19.
Interestingly, reports from several other species, including amphibians, rodents and humans, show co-expression of opsins17,18,20–22. In
humans, for instance, it has been proposed that cones first express S
opsin and later switch to L/M opsin. However, this likely reflects a
developmental process rather than a functional adaptation22.
We identified two major players in the genetic programme for the
transformation of the larval eye to the eyelet. First, EcR acts as a
trigger for both rhodopsin switch and apoptosis. Surprisingly, the
upstream regulators specifying larval photoreceptor-subtype identity, Sal, Svp and Otd, do not contribute to the genetic programme
of sensory plasticity of the rhodopsin switch. Therefore a novel genetic
programme is required for regulating rhodopsin expression in the
eyelet, which likely depends on downstream effectors of EcR.
Second, larval Rh5 and Rh6 photoreceptors respond differently to
ecdysone, either switching rhodopsin expression or undergoing apoptosis. This appears to depend on Sens, which is likely to be required
for the survival of Rh5 photoreceptors. The role of Sens in inhibiting
apoptosis is not unique to this situation: Sens is essential to promote
survival of salivary-gland precursors during embryogenesis23. The
vertebrate homologue of sens, Gfi-1, acts to inhibit apoptosis of
T-cell precursors in haematopoiesis and cochlear hair cells of the
inner ear24,25. Thus the anti-apoptotic function of Sens/Gfi-1 may
be a general property of this molecule.
Ecdysone acts in remodelling neurons during metamorphosis19,20,22,. In c-neurons of the mushroom body, a structure involved
in learning and memory, ecdysone is required for the pruning of
larval processes26. Similarly, dendrites of C4da sensory neurons
undergo large-scale remodelling that depends on ecdysone signalling27. Interestingly, in the moth Manduca, ‘lateral neurosecretory
cells’ express cardio-acceleratory peptide 2, which is switched off in
response to ecdysone before expression of the neuropeptide bursicon
is initiated in the adult28.
The transformation of larval blue-sensitive photoreceptors to
green-sensitive photoreceptors of the eyelet reveals an unexpected
example of sensory plasticity by switching rhodopsin gene expression
in functional, terminally differentiated sensory neurons.
METHODS SUMMARY
Drosophila strains and genetics. The following fly strains were used: yw122; soGal4, rh5-Gal4, rh6-Gal4, rh5-GFP, rh6-GFP, rh5-lacZ otduvi, otd-Gal4 (T. Cook,
personal communication), UAS-sens29, GMR-Gal4 (Bloomington Drosophila
Stock Center), UAS-EcRRNAi, svpH162-LacZ, svp724-Gal4 (Kyoto Stock Center),
sal-Gal4, UAS-EGFRdn, UAS-melt30, UAS-EcRDN, (UAS-EcRDN for isoforms A,
B1 and B2 all gave comparable results; UAS-EcRDN-B2 is used in the figures),
73EcRE-lacZ14, UAS-EcRRNAi, UAS-H2B::YFP (anti-GFP antibody/biogenesis
recognizes the YFP antigen), UAS-svp, UAS-salm, UAS-hid, UAS-rpr, UASGFP, Act-FRT . STOP . FRT-nlacZ, UAS-lacZ, UAS-p35, Tub-Gal80ts
(Bloomington).
Immunohistochemistry and preparation of larval and adult specimens.
Primary antibodies were rabbit anti-Rh6 1:10,000, mouse anti-Rh5 1:20, rat
anti-Elav 1:30 (Developmental Studies Hybridoma Bank (DSHB)), mouse
anti-EcR (DSHB, antibodies against EcR-A, EcR-B1 and EcR-B2 gave comparable results; anti-EcRcommon is used in all figures), sheep anti-GFP (Biogenesis),
rabbit anti-Sal 1:200, guinea pig anti-Sens 1:1000, mouse anti-Svp 1:1000, mouse
anti-Pros 1:50 (DSHB), mouse anti-Chp 1:10 (DSHB), mouse anti-Pdf 1:30
(DSHB), rat anti-Otd 1:200 and mouse anti-betaGAL 1:20 (DSHB).
Secondary antibodies used for confocal microscopic analysis were Alexa-488,
Alexa-555 and Alexa-647 generated in goat (Molecular Probes, Invitrogen), all at
1:300–1:500 dilution. Specimens were mounted in Vectashield H-1000 (Vector).
Dissection for staining of the larval eye was performed as previously described1.
Laser confocal microscopy and image processing. A Leica TCS SP laser confocal microscope was used. Optical sections ranged from 0.2 to 2 mm recorded in
line average mode with picture size of 512 pixels 3 512 pixels, or 1,024 pixels 3
1,024 pixels. Captured images from optical sections were arranged and processed
using Leica Confocal Software and ImageJ, and imported into Adobe Photoshop.
Received 28 February; accepted 9 May 2008.
Published online 25 June 2008.
1.
2.
Sprecher, S. G., Pichaud, F. & Desplan, C. Adult and larval photoreceptors use
different mechanisms to specify the same rhodopsin fates. Genes Dev. 21,
2182–2195 (2007).
Helfrich-Forster, C. et al. The extraretinal eyelet of Drosophila: development,
ultrastructure, and putative circadian function. J. Neurosci. 22, 9255–9266
(2002).
4
©2008 Macmillan Publishers Limited. All rights reserved
LETTERS
NATURE
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Veleri, S., Rieger, D., Helfrich-Forster, C. & Stanewsky, R. Hofbauer-Buchner
eyelet affects circadian photosensitivity and coordinates TIM and PER expression
in Drosophila clock neurons. J. Biol. Rhythms 22, 29–42 (2007).
Mealey-Ferrara, M. L., Montalvo, A. G. & Hall, J. C. Effects of combining a
cryptochrome mutation with other visual-system variants on entrainment of
locomotor and adult-emergence rhythms in Drosophila. J. Neurogenet. 17, 171–221
(2003).
Malpel, S., Klarsfeld, A. & Rouyer, F. Larval optic nerve and adult extra-retinal
photoreceptors sequentially associate with clock neurons during Drosophila brain
development. Development 129, 1443–1453 (2002).
Yasuyama, K. & Meinertzhagen, I. A. Extraretinal photoreceptors at the
compound eye’s posterior margin in Drosophila melanogaster. J. Comp. Neurol. 412,
193–202 (1999).
Mazzoni, E. O., Desplan, C. & Blau, J. Circadian pacemaker neurons transmit and
modulate visual information to control a rapid behavioral response. Neuron 45,
293–300 (2005).
Busto, M., Iyengar, B. & Campos, A. R. Genetic dissection of behavior: modulation
of locomotion by light in the Drosophila melanogaster larva requires genetically
distinct visual system functions. J. Neurosci. 19, 3337–3344 (1999).
Hassan, J., Iyengar, B., Scantlebury, N., Rodriguez Moncalvo, V. & Campos, A. R.
Photic input pathways that mediate the Drosophila larval response to light and
circadian rhythmicity are developmentally related but functionally distinct. J.
Comp. Neurol. 481, 266–275 (2005).
Daniel, A., Dumstrei, K., Lengyel, J. A. & Hartenstein, V. The control of cell fate in
the embryonic visual system by atonal, tailless and EGFR signaling. Development
126, 2945–2954 (1999).
Tahayato, A. et al. Otd/Crx, a dual regulator for the specification of ommatidia
subtypes in the Drosophila retina. Dev. Cell 5, 391–402 (2003).
Kimura, H. Histone dynamics in living cells revealed by photobleaching. DNA
Repair 4, 939–950 (2005).
Talbot, W. S., Swyryd, E. A. & Hogness, D. S. Drosophila tissues with different
metamorphic responses to ecdysone express different ecdysone receptor
isoforms. Cell 73, 1323–1337 (1993).
Kozlova, T. & Thummel, C. S. Essential roles for ecdysone signaling during
Drosophila mid-embryonic development. Science 301, 1911–1914 (2003).
Devarakonda, S., Harp, J. M., Kim, Y., Ozyhar, A. & Rastinejad, F. Structure of the
heterodimeric ecdysone receptor DNA-binding complex. EMBO J. 22,
5827–5840 (2003).
Cheng, C. L. & Flamarique, I. N. Chromatic organization of cone photoreceptors in
the retina of rainbow trout: single cones irreversibly switch from UV (SWS1) to
blue (SWS2) light sensitive opsin during natural development. J. Exp. Biol. 210,
4123–4135 (2007).
Cheng, C. L., Flamarique, I. N., Harosi, F. I., Rickers-Haunerland, J. & Haunerland,
N. H. Photoreceptor layer of salmonid fishes: transformation and loss of single
cones in juvenile fish. J. Comp. Neurol. 495, 213–235 (2006).
Cheng, C. L. & Novales Flamarique, I. Opsin expression: new mechanism for
modulating colour vision. Nature 428, 279 (2004).
Mazzoni, E. O., Desplan, C. & Celik, A. ‘One receptor’ rules in sensory neurons.
Dev. Neurosci. 26, 388–395 (2004).
20. Applebury, M. L. et al. The murine cone photoreceptor: a single cone type
expresses both S and M opsins with retinal spatial patterning. Neuron 27, 513–523
(2000).
21. Makino, C. L. & Dodd, R. L. Multiple visual pigments in a photoreceptor of the
salamander retina. J. Gen. Physiol. 108, 27–34 (1996).
22. Xiao, M. & Hendrickson, A. Spatial and temporal expression of short, long/
medium, or both opsins in human fetal cones. J. Comp. Neurol. 425, 545–559
(2000).
23. Chandrasekaran, V. & Beckendorf, S. K. Senseless is necessary for the survival of
embryonic salivary glands in Drosophila. Development 130, 4719–4728 (2003).
24. Hock, H. et al. Intrinsic requirement for zinc finger transcription factor Gfi-1 in
neutrophil differentiation. Immunity 18, 109–120 (2003).
25. Wallis, D. et al. The zinc finger transcription factor Gfi1, implicated in
lymphomagenesis, is required for inner ear hair cell differentiation and survival.
Development 130, 221–232 (2003).
26. Zheng, X. et al. TGF-b signaling activates steroid hormone receptor expression
during neuronal remodeling in the Drosophila brain. Cell 112, 303–315 (2003).
27. Kuo, C. T., Jan, L. Y. & Jan, Y. N. Dendrite-specific remodeling of Drosophila
sensory neurons requires matrix metalloproteases, ubiquitin-proteasome, and
ecdysone signaling. Proc. Natl Acad. Sci. USA 102, 15230–15235 (2005).
28. Loi, P. K. & Tublitz, N. J. Hormonal control of transmitter plasticity in insect
peptidergic neurons. I. Steroid regulation of the decline in cardioacceleratory
peptide 2 (CAP2) expression. J. Exp. Biol. 181, 175–194 (1993).
29. Nolo, R., Abbott, L. A. & Bellen, H. J. Senseless, a Zn finger transcription factor, is
necessary and sufficient for sensory organ development in Drosophila. Cell 102,
349–362 (2000).
30. Mikeladze-Dvali, T. et al. The growth regulators warts/lats and melted interact in
a bistable loop to specify opposite fates in Drosophila R8 photoreceptors. Cell 122,
775–787 (2005).
Supplementary Information is linked to the online version of the paper at
www.nature.com/nature.
Acknowledgements We thank H. Bellen, A. H. Brand, S. Britt, T. Cook, the
Developmental Studies Hybridoma Bank, F. Hirth, the Kyoto Stock Center,
K. Matthews, M. Mlodzik, B. Mollerau, F. Pichaud, H. Reichert, C. Thummel and
J. Urban for fly stocks and antibodies. We also thank J. Blau, R. J. Johnston, A. Keene
and D. Vasiliauskas for discussion and comments on the manuscript. This work
was funded by grant EY013010 from the National Eye Institute/National Institutes
of Health to C.D., the Swiss National Science Foundation, the Novartis Foundation
and the Janggen-Pöhn Stiftung (to S.G.S.) and conducted in a facility constructed
with the support of a Research Facilities Improvement Grant C06 RR-15518-01
from the National Center for Research Resources, National Institutes of Health.
Author Contributions S.G.S. performed the experimental work and analysed the
data. C.D. and S.G.S. designed the experiments and wrote the paper.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. Correspondence and requests for materials should be
addressed to C.D. ([email protected]).
5
©2008 Macmillan Publishers Limited. All rights reserved