www.sciencemag.org/cgi/content/full/340/6134/875/DC1

Supplementary Materials for

ATAXIN-2 Activates PERIOD Translation to Sustain Circadian Rhythms in Drosophila

Chunghun Lim and Ravi Allada†

*Corresponding author. E-mail: r-allada@northwestern.edu

Published 17 May 2013, Science 340, 875 (2013) DOI: 10.1126/science.1234785

This PDF file includes:

Materials and Methods Figs. S1 to S18 Table S1 References

2

Materials and Methods

Plasmids

MS2 reporter plasmids, pAc/FLUC/per 3’UTR/6xBS and pAc/RLUC, have been described previously (6). Full-length or deletion mutants of tyf or Atx2 cDNAs were inserted into 1) pAc/MS2-V5 to express C-terminal MS2-V5 fusion proteins in S2 cells, 2) pAc/3xF to express C-terminal 3xFLAG fusion proteins in S2 cells, 3) a modified pUAS-C5 with a C-terminal 3xFLAG tag to generate UAS-TYF-3F, UAS-TYFΔC5-3F, UAS-ATX2-3F, and UAS-ATX2ΔPAM2-3F transgenic flies. For the generation of a site-specific transgenic fly, TYFΔC5 cDNA was inserted into a modified pUAST with attB site and C-terminal V5-tag.

RNA tethering assay

S2 cells on 24-well plates were cotransfected with 5 ng of Ac/FLUC/per3’UTR/6xBS, 5 ng of Ac/RLUC, and 90 ng of Ac/MS2-V5 (or its derivatives) by Effectene transfection reagent according to manufacturer’s instructions (Qiagen) or a standard calcium precipitation method. The amounts of plasmid DNA and transfection reagents were proportionally scaled up for the transfection on 6-well plates. For RNAi experiments, dsRNAs were synthesized by MEGAscript RNAi kit according to the manufacturer’s instructions (Invitrogen). Four or 15 μg of dsRNA was incubated with S2 cells on 24-well or 6-well plates, respectively. At 48 h later, dsRNA-treated S2 cells were cotransfected with MS2 reporters and MS2-fusion expression vector. Cells were harvested at 48 h after transfection and dual luciferase assays were performed according to the manufacturer’s instructions (Promega). The rest of cell extracts were resolved by SDS-PAGE and immunoblotted with specific antibodies to confirm the knockdown of target proteins.

Immunoprecipitation (IP)

S2 cells on 100 mm dishes were transfected with expression vector for 3xFLAG-tagged proteins. Cells were harvested at 48 h after transfection, washed in PBS, and resuspended in T150 lysis buffer (25 mM Tris-Cl pH 7.5, 150 mM NaCl, 10% glycerol, 1 mM EDTA, 1 mM dithiothreitol, 0.5% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, protease inhibitors). For the IP from transgenic flies, ~200 fly heads were homogenized in T300 lysis buffer (25 mM Tris-Cl pH 7.5, 300 mM NaCl, 10% glycerol, 1 mM EDTA, 1 mM dithiothreitol, 0.5% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, protease inhibitors). After incubation at 4°C for 20 min, the salt concentration was adjusted to 200 mM. Extracts were clarified by centrifugation and incubated with pre-equilibrated anti-FLAG M2 affinity gel (Sigma) for 1.5 h at 4°C. Where indicated, extracts were pre-treated with 0.5 mg/ml of RNase A at room temperature for 5 min before the IP. The beads were washed four times with the same buffer. Bound proteins were eluted by boiling in 1x SDS sample buffer, resolved by SDS-PAGE and immunoblotted with specific antibodies. For mass spectrometric analyses (LC-MS/MS), TYF IP complexes were resolved by SDS-PAGE and visualized by SimplyBlue SafeStain (Invitrogen). TYF-C5-specific protein bands were excised from the gel and

3

subjected to in-gel trypsin digestion before the LC-MS/MS and data analyses by Proteomics Core at Northwestern University. Alternatively, TYF IP complexes were eluted by incubating with 150 ng/μl of 3xFLAG peptide and subjected to in-solution trypsin digestion.

Antibodies

Mouse anti-AGO1 (Argonaute-1) and anti-GW182 (gifts from Dr. Siomi), mouse anti-FLAG (Sigma), mouse anti-Me31B (maternal expression at 31B) (a gift from Dr. Nakamura), mouse anti-NONA (no on or off transient A) (a gift from Dr. Saumweber), rabbit anti-ATX2 (a gift from Drs. Pallanck and Ramaswami), rabbit anti-LARK (a gift from Dr. McNeil), rabbit anti-dPABP (a gift from Dr. Sonenberg), rabbit anti-TYF (a gift from Dr. Choe), rat anti-GE-1 and anti-TRAL (trailer hitch) (gifts from Dr. Izaurralde), and rat anti-HA (Roche) were used in our protein analyses.

RNA immunoprecipitation (RIP) and quantitative RT-PCR

Two hundreds of fly heads were homogenized in a HEPES RIP buffer (20 mM HEPES pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM dithiothreitol, 0.5% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 250 U/ml SUPERase-In, protease inhibitors). After incubation at 4°C for 20 min, head extracts were clarified by centrifugation twice and incubated with pre-equilibrated anti-FLAG M2 affinity gel (Sigma) for 1.5 h at 4°C. The beads were washed five times with the same buffer. Bound RNAs were extracted with Trizol reagent (Invitrogen), further purified by RNeasy Micro Kit (Qiagen), and then reverse transcribed by M-MLV reverse transcriptase (Promega) according to the manufacturer’s instructions. cDNAs converted from input and IP RNAs were quantitatively analyzed by SsoAdvanced SYBR Green Supermix (Bio-Rad) using CFX384 Touch real-time PCR detection system (Bio-Rad).

Fly stocks and behavioral assays

Pdf-GAL4, tim-GAL4-62, and UAS-per16 transgenic flies were described previously (16-18). A strongly hypomorphic Atx2[06490] mutant allele (Bloomington stock # 11688, denoted as Atx2[0] in figs. S6 and S13) is homozygous lethal and has a P element insertion in a 5’ untranslated region of Atx2 isoforms causing ATX2 protein expression undetectable in homozygous mutant clones (19). Two UAS-Atx2 RNAi lines were obtained from Bloomington Drosophila stock center (TRiP line, Bloomington stock #36114) and Vienna Drosophila RNAi center (KK line, #108843). Several independent germ-line transformants were established from w1118 embryos injected with each of UAS transgenic constructs (BestGene Inc.). UAS-TYFΔC-V5 transgene was inserted into the same attP40 landing site used for the UAS-TYF-V5 transgenic fly in our previous study (6) to pair wisely compare their difference. All flies were reared with standard cornmeal-yeast-agar medium at 25°C under LD (12-h light/12-h dark) cycles. Locomotor activity of individual male flies was measured using Drosophila Activity Monitors (Trikinetics). Flies were first entrained by LD cycles for 5 days and then transferred to constant dark (DD) cycles for a week to monitor circadian behaviors in the free-running condition. The period, power (P), and significance (S) values were calculated by chi-squared

4

periodogram (29) with the confidence set to 0.05 using ClockLab analysis software (Actimetrics). The power of rhythmicity in DD behavior was measured by P-S values in individual flies and averaged for each genotype and reflects rhythm amplitude.

Immunostaining

Whole-mount immunostaining in adult fly brains was performed as described previously (6). For primary antibodies, we used rabbit anti-PER (a gift from Dr. Rosbash), guinea pig anti-TIM (a gift from Dr. Choe), guinea pig anti-VRI (a gift from Dr. Hardin), rabbit anti-ATX2 (a gift from Drs. Pallanck and Ramaswami) and mouse anti-PDF (Developmental Studies Hybridoma Bank) antibodies. For fluorescence-conjugated secondary antibodies, anti-rabbit IgG Alexa Fluor 488 and 594, anti-guinea pig IgG Alexa Fluor 488 and 594, and anti-mouse IgG Alexa Fluor 647 antibodies (Jackson Immuno Research Laboratories, Inc.) were used. Images were obtained using confocal laser-scanning microscopes (Nikon C1). For the quantitative analysis, signal intensity from each group of clock cells was quantified using ImageJ software as described previously (6).

PABP

AGO1

TYF-3HA

vecto

r

AT

X2

-3F

AT

X2Δ

C-3

F

AT

X2Δ

N-3

F

AT

X2

-C-3

F

AT

X2

-N-3

F

TRAL

vecto

r

AT

X2

-3F

AT

X2Δ

C-3

F

AT

X2Δ

N-3

F

AT

X2

-C-3

F

AT

X2

-N-3

F

input (4.5%) anti-FLAG IP A

(kDa) 200 116

97 66

45

vecto

r

AT

X2

-3F

AT

X2Δ

C-3

F

AT

X2

-C-3

F

AT

X2

-N-3

F

AT

X2Δ

N-3

F

B

ATX2ΔN

ATX2-N

ATX2-C

ATX2ΔC

ATX2ΔPAM2

ATX2

299

1084

854

1

Δ855-869

LsmA

D

PAM2

298

785

Fig. S1. N-terminal ATX2 region containing LsmA interacts with TYF. (A) Schematic diagram of ATX2

deletion constructs used in mapping TYF- and PABP-interacting domains. LsmA, Lsm/Lsm-associated;

PAM2, PABP-interacting motif 2. (B) S2 cells were cotransfected with expression vectors for 3xHA-

tagged TYF and each of 3xFLAG-tagged ATX2 domains. 3xFLAG-tagged ATX2 domains (left panel) were

immunoprecipitated from S2 cell extracts. Bound proteins were probed with specific antibodies (right

panel)

ATX2-3F

AGO1

TYF-3HA

PABP

GST-3F

ATX2-3F

ATX2ΔPAM2-3F

RNase A (0.5 mg/ml) + + +

+

+

+

+

+

+

+

+

+

input (4.5%) anti-FLAG IP

GST-3F

A

tim

>

tim

>A

TX

2-3

F

tim

>Δ

PA

M2-3

F

input(4.5%) anti-FLAG IP

tim

>

tim

>A

TX

2-3

F

tim

>Δ

PA

M2-3

F

PABP

TYF

AGO1

FLAG

B

Fig. S2. TYF-ATX2 interaction is independent of PABP and RNA. (A) S2 cells were

cotransfected with expression vectors for 3xHA-tagged TYF and each of 3xFLAG-tagged

glutathione S-transferase (GST, negative control), wild-type ATX2 or PAM2 domain

deletion (ATX2ΔPAM2). 3xFLAG-tagged proteins were immunoprecipitated from S2 cell

extracts and bound proteins were probed with specific antibodies. Where indicated, cell

extracts were treated with 0.5 mg/ml of RNase A prior to the immunoprecipitation (IP) with

anti-FLAG antibody. (B) 3xFLAG-tagged ATX2 and ATX2ΔPAM2 were expressed in all

clock neurons by tim-Gal4 (tim>). Adult flies were collected at ZT15 during a light-dark

cycle (lights-on at ZT0; lights-off at ZT12). Head extracts were immunoprecipitated with

anti-FLAG affinity gels and bound proteins were analyzed as in (A).

22

24

26

28

30

32

***

Fig. S3. Long-period rhythms caused by TYFΔC5 overexpression are rescued by PER overexpression.

Epitope-tagged TYFΔC5 lacking the ATX2-binding domain (ΔC5-V5 or ΔC5-3F) and/or PER was

overexpressed in PDF neurons of wild-type flies by Pdf-Gal4 (Pdf>). Averaged actograms during light-dark

(LD) and constant dark (DD) cycles are double-plotted at the top. Genotypes, number of flies tested, period

(hour) and power of rhythmicity (as measured by power (P) – significance (S) values) under DD cycles are

shown with their statistical analyses at the bottom. Data represent average+/-SEM (n=23-61). n.s., not

significant at the 0.05 level; **PΔC5-V5,PER

n=34, 25.3+/-0.2 h

P-S=66.7+/-6.7

Pdf>PER

n=28, 24.6+/-0.1 h

P-S=97.3+/-10.8

Pdf>

n=35, 24.7+/-0.1 h

P-S=99.1+/-6.1

LD

D

D

LD

D

D

Pdf>ΔC5-3F #7

n=27, 30.8+/-0.3 h

P-S=53.2+/-4.4

Pdf>ΔC5-3F#7,PER

n=23, 24.5+/-0.1 h

P-S=72.3+/-7.3

0

20

40

60

80

100

120

Period (

h)

***

Pow

er

of R

hyth

mic

ity

(P-S

)

n.s.

n.s.

+/+ u-PER/+ +/+ u-PER/+

Pdf-Gal4/+ Pdf-Gal4/

u-ΔC5-V5

Pdf-Gal4/

u-ΔC5-3F#7

Pdf-Gal4/+ Pdf-Gal4/

u-ΔC5-V5

Pdf-Gal4/

u-ΔC5-3F#7

n.s.

***

***

n.s.

**

Fig. S4. Targeted expression of Atx2 RNAi transgene decreases endogenous ATX2 protein levels in

transgenic flies. (A) Atx2 RNAi transgene (TRiP) was overexpressed in all clock neurons by tim-Gal4 along

with RNAi-enhancing dcr2 transgene (tim>dcr2). Adult fly brains were dissected at ZT20 under LD cycle

(lights-on at ZT0; lights-off at ZT12) and immunostained with anti-ATX2 (green) and anti-VRI (red)

antibodies. VRI-expressing large ventral lateral neurons are indicated with dotted circles. Similar result was

obtained using the other Atx2 RNAi transgene (KK line, not shown). (B) Two independent Atx2 RNAi

transgenes were overexpressed in eyes (by GMR-Gal4) or all clock neurons (by tim>dcr2). Flies were

entrained under LD cycles and collected at the indicated time-points. Head extracts were resolved by SDS-

PAGE and immunoblotted with specific antibodies shown on the left.

PER

TIM

TYF

ATX2

PABP

ZT

3

ZT

19

ZT

3

ZT

19

ZT

3

ZT

19

GM

R>

GM

R>

Atx

2 R

NA

i (K

K)

GM

R>

Atx

2 R

NA

i (T

RiP

)

PER

TIM

TYF

ATX2

PABP

ZT

3

ZT

19

ZT

3

ZT

19

ZT

3

ZT

19

tim

>d

cr2

tim

>d

cr2

Atx

2 R

NA

i (K

K)

tim

>d

cr2

Atx

2 R

NA

i (T

RiP

)

AT

X2

VR

I A

TX

2

tim>dcr2

Atx2 RNAi (TRiP) tim>dcr2

B A

Fig. S5. Wild-type ATX2 or PER overexpression rescues long-period rhythms by Atx2 knockdown in

PDF neurons. Wild-type ATX2 or PER was co-overexpressed along with Atx2 RNAi transgene (KK)

in PDF neurons by Mz520-Gal4. The period (hour) and power of rhythmicity (as measured by power

(P) – significance (S) values) of their locomotor rhythms under constant dark cycles are shown with

their statistical analyses. Data represent average+/-SEM (n=24-68). n.s., not significant at the 0.05

level; *P

22

23

24

25

26

27

28

0

20

40

60

80

100

120

140

** ***

n.s.

n.s.

***

***

Period (

h)

Mz520/+ Mz520/+;

Atx2[0]/+

Mz520/

Atx2RNAi

(KK)

Mz520/

Atx2RNAi

(KK);

Atx2[0]/+

p

0

20

40

60

80

100

120

140

160

22

23

24

25

26

27

28

Fig. S7. Strong Atx2 knockdown in PDF neurons causes arrhythmic circadian behaviors. Atx2 RNAi

transgene (KK) was overexpressed in PDF neurons using Pdf-Gal4 along with RNAi-enhancing dcr2

transgene (Pdf>dcr2). The period (hour) and power of rhythmicity (as measured by power (P) –

significance (S) values) of their locomotor rhythms under constant dark cycles are shown with their

statistical analyses. Data represent average+/-SEM (n=41-66). n.s., not significant at the 0.05 level;

***Pdcr2/+ Pdf>dcr2/

Atx2RNAi

(KK)

Atx2RNAi

(KK)/+

Pow

er

of R

hyth

mic

ity

(P-S

) n.s.

***

Pdf>dcr2/+ Pdf>dcr2/

Atx2RNAi

(KK)

Atx2RNAi

(KK)/+

0

20

40

60

80

100

120

140

Fig. S8. Atx2 knockdown in all pacemaker neurons causes severe arrhythmicity in circadian

behaviors. Each of two independent Atx2 RNAi transgenes (KK or TRiP) was overexpressed in

all clock neurons by tim-Gal4 driver along with RNAi-enhancing dcr2 transgene (tim>dcr2).

Averaged actograms during light-dark (LD) and constant dark (DD) cycles are double-plotted.

Genotypes, number of flies tested, period (hour) and power of rhythmicity (as measured by

power (P) – significance (S) values) under DD cycles are shown at the top. Statistical analyses

on the power of rhythmicity in controls and Atx2 knockdown flies are shown at the bottom.

Data represent average+/-SEM. ***Pdcr2

n=32, 24.3+/-0.1h

P-S=96.5+/-7.3

tim>dcr2,Atx2 RNAi(TRiP)

n=31, 25.7+/-1.4h

P-S=16.7+/-4.6

tim>dcr2,Atx2 RNAi (KK)

n=20, 100% arrhythmic

P-S=2.2+/-0.9

LD

D

D

Pow

er

of R

hyth

mic

ity

(P-S

)

tim-Gal4/+;

u-dcr2/+

tim-Gal4/

Atx2RNAi(KK);

u-dcr2/+

tim-Gal4/+;

u-dcr2/

Atx2RNAi

(TRiP)

Atx2RNAi(KK)/+;

u-dcr2/+

u-dcr2/

Atx2RNAi

(TRiP)

***

tim>dcr2 tim>dcr2,Atx2 RNAi (KK) tim>dcr2,Atx2 RNAi (TRiP)

0

20

40

60

80

100

120

0

20

40

60

80

100

120

0

20

40

60

80

100

120

LNd l-LNv s-LNv PDF(-) s-LNv

PE

R le

ve

ls (

%)

TIM

le

ve

ls (

%)

VR

I le

ve

ls (

%)

A

Fig. S9. Atx2 knockdown in all pacemaker neurons strongly dampens PER cycling.

(A) Each of two independent Atx2 RNAi transgenes (KK or TRiP) was overexpressed

in all clock neurons by tim-Gal4 along with RNAi-enhancing dcr2 transgene

(tim>dcr2). Adult fly brains were dissected at different time-points under a LD cycle

(lights-on at ZT0; lights-off at ZT12) and immunostained with anti-PER, anti-TIM, or

anti-VRI (VRILLE) antibodies. PER, TIM, and VRI levels in each group of lateral

neurons (LNs) were quantified and normalized to the values of the Gal4 control at

peak time-points (tim>dcr2, set as 100%) as in Figs. 2B and 2C. Data represent

average+/-SEM (n=8-11). #Pdcr2) at each time-point

as determined by one-way ANOVA, Tukey post-hoc test. LNd, dorsal LN; l-LNv, large

ventral LN; s-LNv, small ventral LN; PDF(-), PDF-negative.

#

#

#

# # #

#

#

#

# #

#

# #

#

#

#

# #

#

#

#

#

PE

R

TIM

ZT0 ZT4 ZT8 ZT20 ZT16

Me

rge

PD

F

PE

R

TIM

Me

rge

PD

F

B

PE

R

TIM

Me

rge

PD

F

tim

>d

cr2

tim

>d

cr2

Atx

2 R

NA

i (K

K)

tim

>d

cr2

Atx

2 R

NA

i (T

RiP

)

Fig. S9 (continued). (B) Representative confocal images

of small ventral LNs were obtained at each time-point by

co-staining with anti-PER, anti-TIM, and anti-PDF

antibodies.

tim

>dcr2

tim

>dcr2

,Atx

2 R

NA

i (K

K)

tim

>dcr2

,Atx

2 R

NA

i (T

RiP

)

ZT17 ZT13 ZT9 ZT21 ZT1 ZT5

VR

I V

RI

PD

F

VR

I V

RI

PD

F

VR

I V

RI

PD

F

*

LNd

l-LNv

s-LNv

LNd

l-LNv

s-LNv

*

LNd

l-LNv

s-LNv

*

*

LNd LNd LNd

s-LNv

l-LNv

l-LNv

s-LNv

s-LNv

l-LNv

l-LNv

* l-LNv

s-LNv

LNd LNd LNd

l-LNv *

s-LNv

l-LNv

s-LNv * l-LNv

*

s-LNv

l-LNv

LNd

LNd LNd

LNd LNd

l-LNv

l-LNv

s-LNv

s-LNv

C

Fig. S9 (continued). (C) Representative confocal images of LNs were obtained at each time-point

by co-staining with anti-VRI and anti-PDF antibodies. LNd, dorsal LN; l-LNv, large ventral LN; s-LNv, small ventral LN. Asterisk indicates PDF-negative s-LNv.

0%

20%

40%

60%

80%

100%

0%

20%

40%

60%

80%

100%

%D

ors

al P

roje

ction

%P

oste

rior

Op

tic T

ract

wild-type

modest outgrowth

extra branching/abnormal targeting

wild-type

extra branching

disconnected

+

Atx

2 R

NA

i (K

K)

Pdf>dcr2 tim>dcr2

DP

POT

B

A

Fig. S10. Atx2 knockdown in PDF neurons does not affect axonal projections from ventral lateral

neurons (LNs). (A) Atx2 RNAi transgene (KK) was overexpressed in PDF neurons by Pdf-Gal4 along

with RNAi-enhancing dcr2 transgene (Pdf>dcr2) or in all clock neurons by tim-Gal4 (tim>dcr2). Adult fly

brains were dissected and immunostained with anti-PDF antibody. Representative, z-stacked images

are shown. DP, dorsal projection from small ventral LNs; POT, posterior optic tract from large ventral

LNs. (B) The abnormality in PDF projections from large and small ventral LNs was scored from each

brain hemisphere (n=78-136). The percentage of wild-type and mutant projections were calculated per

each genotype.

22

23

24

25

26

0

20

40

60

80

100

120

140

160

Mz520/+ Atx2RNAi

(KK)/+

Mz520>

Atx2RNAi#/+

tyf[Δ];

Mz520/+

tyf[Δ];

Atx2RNAi

(KK)/+

tyf[Δ];

Mz520>

Atx2RNAi#/+

n.s.

Pow

er

of

Rhyth

mic

ity (

P-S

)

Mz520/+ Atx2RNAi

(KK)/+

Mz520>

Atx2RNAi#/+

tyf[Δ];

Mz520/+

tyf[Δ];

Atx2RNAi

(KK)/+

tyf[Δ];

Mz520>

Atx2RNAi#/+

Peri

od

(h

)

Fig. S11. Atx2 knockdown and tyf mutation do not show additive effects on period and rhythmicity.

Atx2 RNAi transgene (KK) was overexpressed in PDF neurons using Mz520-Gal4 in a wild-type or

tyf mutant background (tyf[Δ]). The period (hour) and power of rhythmicity (as measured by power

(P) – significance (S) values) of their locomotor rhythms under constant dark cycles are shown with

their statistical analyses. Data represent average+/-SEM (n=22-47). n.s., not significant at the 0.05

level; **P

12 13 14 15 16 17

Fig. S12. Atx2 knockdown and tyf mutation do not show additive effects on the delayed circadian

phases of locomotor activity peaks. (A) Atx2 RNAi transgene (KK) was overexpressed in PDF

neurons using Mz520-Gal4 in a wild-type or tyf mutant background (tyf[Δ]). Averaged actograms

during the first three cycles of constant dark (DD1-DD3) are shown (n=22-47). Mz520>Atx2RNAi#, a

recombinant transgenic line of Mz520-Gal4 and UAS-Atx2RNAi (KK) transgenes. 50% offsets in

evening activity peaks during the second DD cycle were indicated by arrows. (B) The circadian

phases of 50% offset in evening activity peaks during the second DD cycle were calculated and

averaged from three groups of flies per a given genotype. Data represent average+/-SEM (n=3). n.s.,

not significant at the 0.05 level; **PAtx2RNAi#/+

Phase of 50% Offset in

Evening Activity Peaks (CT)

** *** *** ***

n.s

.

p<

0.0

02

DD2 DD3 DD1

Mz520/+ Atx2 RNAi(KK)/+ Mz520>Atx2RNAi#/+

Mz520/+ tyf[Δ]; Mz520/+

tyf[Δ]; Atx2 RNAi(KK)/+ tyf[Δ]; Mz520/+ tyf[Δ]; Mz520>Atx2RNAi#/+

B

A

n.s

.

Fig. S13. A heterozygous mutation for a strongly hypomorphic Atx2 allele enhances period

lengthening by TYFΔC5 overexpression. V5-tagged TYFΔC5 lacking the ATX2-binding domain

(ΔC5-V5) was overexpressed in PDF neurons of wild-type or a heterozygous Atx2 mutant flies

(Atx2[0]/+) by Pdf-Gal4. The period (hour) and power of rhythmicity (as measured by power (P) –

significance (S) values) under DD cycles are shown with their statistical analyses. Data represent

average+/-SEM (n=25-61). n.s., not significant at the 0.05 level; ***P

Fig. S14. Time-dependent RNA enrichment by ATX2 immunoprecipitation (IP) from fly

head extracts. 3xFLAG-tagged ATX2 was expressed in all clock neurons of wild-type flies

by tim-Gal4. Adult flies were collected at ZT3 or ZT15 under a LD cycle (lights-on at ZT0;

lights-off at ZT12) and their head extracts were immunoprecipitated with anti-FLAG affinity

gels. Bound RNAs were analyzed by quantitative real-time RT-PCR using each gene-

specific primer set in triplicate. (A) RNA enrichment fold was calculated by first subtracting

RNA levels in control IP (tim>) from those in ATX2-3xFLAG IP (tim>ATX2-3F) and then

normalizing each subtracted RNA level at ZT15 to ZT3 (except that Clk RNA levels at ZT3

were normalized to ZT15). Data represent average+/-SEM (n=3). n.s., not significant at the

0.05 level; *P

wild-type

ZT

3

ZT

7

ZT

11

ZT

15

ZT

19

ZT

23

ZT

3

ZT

3

ZT

3

ZT

15

ZT

15

ZT

15

PER

TIM

TYF

ATX2

PABP

AGO1

GW182

Fig. S15. ATX2 protein levels in head extracts are not regulated by circadian

clock. Wild-type and clock mutant flies were entrained under LD cycles and

collected at the indicated time-points (lights-on at ZT0; lights-off at ZT12). Head

extracts were resolved by SDS-PAGE and immunoblotted with specific antibodies

shown on the left.

0

20

40

60

80

100

120

140

160

180

n.s. **

*** ***

n.s.

*** ***

n.s. n.s.

n.s. n.s.

n.s.

Pow

er

of R

hyth

mic

ity (

P-S

)

pA

TX

2-3

F #

4

tim

>A

TX

2-3

F #

6

tim

>Δ

PA

M2-3

F #

6

tim

>Δ

PA

M2-3

F #

8

B

A

Fig. S16. Overexpression of a mutant ATX2 protein lacking PABP-interacting motif 2 (ΔPAM2)

causes arrhythmic circadian behaviors. (A) 3xFLAG-tagged wild-type ATX2 (ATX2-3F) or ΔPAM2

(ΔPAM2-3F) was overexpressed in all clock neurons by tim-Gal4 (tim>). Adult head extracts were

prepared from the transgenic flies, resolved by SDS-PAGE, and immunoblotted with specific

antibodies shown on the left. Two independent transgenic flies per each UAS-ATX2-3F or UAS-

ΔPAM2-3F were tested. (B) ATX2-3F or ΔPAM2-3F was overexpressed in PDF neurons by either

Mz520-Gal4 or Pdf-Gal4. The power of rhythmicity (as measured by power (P) – significance (S)

values) of their locomotor rhythms under constant dark cycles is shown with their statistical

analyses. Data represent average+/-SEM (n=24-68). n.s., not significant at the 0.05 level;

**P

PE

R

TIM

ZT0 ZT4 ZT8 ZT20 ZT16

Me

rge

PD

F

PE

R

TIM

Me

rge

PD

F

Mz520>

M

z520>Δ

PA

M2-3

F

Mz520>ΔPAM2-3F

020406080

100120

020406080

100120

Mz520>

LNd l-LNv s-LNv PDF(-) s-LNv

PE

R levels

(%

) T

IM levels

(%

)

B

A

Fig. S17. Overexpression of a mutant ATX2 protein lacking PABP-interacting motif 2

(ΔPAM2) strongly dampens PER cycling. (A) 3xFLAG-tagged ΔPAM2 (ΔPAM2-3F) was

overexpressed in PDF neurons by Mz520-Gal4 (Mz520>). Adult fly brains were dissected

at different time-points under a LD cycle (lights-on at ZT0; lights-off at ZT12) and co-

immunostained with anti-PER, anti-TIM and anti-PDF antibodies. PER and TIM levels in

each group of lateral neurons (LNs) were quantified and normalized to the values of the

Gal4 control at peak time-points (Mz520>, set as 100%) as in Figs. 2B and 2C. Data

represent average+/-SEM (n=8-14). #P) at each time-

point as determined by one-way ANOVA, Tukey post-hoc test. LNd, dorsal LN; l-LNv, large

ventral LN; s-LNv, small ventral LN; PDF(-), PDF-negative. (B) Representative confocal

images of s-LNvs were obtained at each time-point by co-staining with anti-PER, anti-TIM,

and anti-PDF antibodies.

#

# # #

#

Fig. S18. Knockdown of each component in miRNA-induced silencing

complex (miRISC) does not affect RNA-tethered TYF (TYF-MS2) or ATX2

(ATX2-MS2) activation. S2 cells were pre-treated with double-stranded

RNAs for GFP control (dsGFP) or genes indicated at the bottom and

subsequently transfected with firefly luciferase (fluc) reporter containing

MS2-binding sites, renilla luciferase (rluc) reporter, and expression vector

for MS2-fusion protein. FLUC activity was normalized to RLUC activity.

Activation fold was calculated by normalizing to the value with MS2 control

in the presence of the dsGFP control. Data represent average+/-SEM

(n=3-5). n.s., not significant at the 0.05 level; ***P

Genotype na %Rb Period +/- SEM Power +/- SEMc

Pdf-GAL4/+

Table S1. ATX2-binding domain of TYF is required to rescue circadian

behaviors in tyf mutant flies

a n indicates number of flies analyzed b %R indicates percent flies with detectable rhythmicity (P-S>20) c Power is a measure of rhythmic strength d V5-tagged TYF and TYFΔC5 transgenes were inserted into the same genomic locus (attP40).

Pdf-GAL4/UAS-TYF-V5 (attP40)d

Pdf-GAL4/+; UAS-TYF-3F #1/+

Pdf-GAL4/+; UAS-TYF-3F #3/+

Pdf-GAL4/UAS-TYFΔC5-V5 (attP40)

Pdf-GAL4/+; UAS-TYFΔC5-3F #4/+

35 100 24.7 +/- 0.1 99.1 +/- 6.1

33 100 24.9 +/- 0.0 114.1 +/- 8.2

38 92 27.2 +/- 0.2 50.1 +/- 3.4

30 93 26.9 +/- 0.1 75.0 +/- 7.4

61 75 26.8 +/- 0.2 43.8 +/- 4.5

31 71 31.4 +/- 0.4 37.5 +/- 4.5

27 93 30.9 +/- 0.4 49.8 +/- 5.1

27 100 30.8 +/- 0.3 53.2 +/- 4.4

Pdf-GAL4/UAS-TYFΔC5-3F #6

Pdf-GAL4/+; UAS-TYFΔC5-3F #7/+

68 34 23.9 +/- 0.4 21.2 +/- 4.0

33 88 25.2 +/- 0.1 84.2 +/- 8.7

45 87 27.1 +/- 0.1 75.2 +/- 6.6

44 98 26.7 +/- 0.1 92.1 +/- 4.9

46 35 29.3 +/- 1.0 19.0 +/- 2.7

38 42 32.5 +/- 0.7 21.6 +/- 3.3

33 55 30.8 +/- 0.8 28.0 +/- 4.6

39 46 31.4 +/- 0.7 25.9 +/- 4.4

tyfΔ/Y; Pdf-GAL4/+

tyfΔ/Y; Pdf-GAL4/UAS-TYF-V5 (attP40)

tyfΔ/Y; Pdf-GAL4/+; UAS-TYF-3F #1/+

tyfΔ/Y; Pdf-GAL4/+; UAS-TYF-3F #3/+

tyfΔ/Y; Pdf-GAL4/UAS-TYFΔC5-V5 (attP40)

tyfΔ/Y; Pdf-GAL4/+; UAS-TYFΔC5-3F #4/+

tyfΔ/Y; Pdf-GAL4/UAS-TYFΔC5-3F #6

tyfΔ/Y; Pdf-GAL4/+; UAS-TYFΔC5-3F #7/+

39 15 25.5 +/- 1.0 9.4 +/- 1.8

24 88 25.2 +/- 0.1 61.2 +/- 7.6

46 91 27.4 +/- 0.1 67.1 +/- 5.4

31 100 26.5 +/- 0.1 101.0 +/- 8.5

30 33 30.0 +/- 1.1 17.7 +/- 3.1

25 28 33.5 +/- 0.3 15.1 +/- 3.5

32 31 32.7 +/- 0.4 15.9 +/- 2.8

43 42 32.4 +/- 0.3 23.9 +/- 3.8

tyfe/Y; Pdf-GAL4/+

tyfe/Y; Pdf-GAL4/UAS-TYF-V5 (attP40)

tyfe/Y; Pdf-GAL4/+; UAS-TYF-3F #1/+

tyfe/Y; Pdf-GAL4/+; UAS-TYF-3F #3/+

tyfe/Y; Pdf-GAL4/UAS-TYFΔC5-V5 (attP40)

tyfe/Y; Pdf-GAL4/+; UAS-TYFΔC5-3F #4/+

tyfe/Y; Pdf-GAL4/UAS-TYFΔC5-3F #6

tyfe/Y; Pdf-GAL4/+; UAS-TYFΔC5-3F #7/+

Genotype na %Rb Period +/- SEM Power +/- SEMc

+/+

Table S2. Overexpression of a mutant ATX2 protein lacking PABP-interacting

motif 2 (ATX2ΔPAM2) causes arrhythmic circadian behaviors

a n indicates number of flies analyzed b %R indicates percent flies with detectable rhythmicity (P-S>20) c Power is a measure of rhythmic strength

UAS-ATX2-3F #4/+

UAS-ATX2-3F #6/+

UAS- ATX2ΔPAM2-3F #6/+

UAS- ATX2ΔPAM2-3F #8/+

31 100 23.9 +/- 0.1 131.3 +/- 6.5

38 100 23.7 +/- 0.1 120.8 +/- 7.8

32 100 23.9 +/- 0.1 144.3 +/- 6.4

35 100 23.8 +/- 0.1 136.6 +/- 7.7

39 100 23.5 +/- 0.0 159.4 +/- 6.9

68 99 23.8 +/- 0.1 116.0 +/- 5.3

29 100 24.3 +/- 0.1 81.4 +/- 7.5

24 96 23.8 +/- 0.1 87.0 +/- 8.3

32 38 24.1 +/- 0.7 18.7 +/- 3.1

25 28 24.0 +/- 0.5 12.8 +/- 2.5

Mz520-GAL4/+

Mz520-GAL4/UAS-ATX2-3F #4

Mz520-GAL4/+; UAS-ATX2-3F #6/+

Mz520-GAL4/ UAS-ATX2ΔPAM2-3F #6

Mz520-GAL4/+; UAS-ATX2ΔPAM2-3F #8/+

35 100 24.7 +/- 0.1 99.1 +/- 6.1

32 91 24.5 +/- 0.1 77.4 +/- 7.1

28 93 24.9 +/- 0.1 69.6 +/- 6.5

33 45 23.3 +/- 0.2 26.0 +/- 5.9

32 31 24.2 +/- 0.8 17.7 +/- 3.4

Pdf-GAL4/+

Pdf-GAL4/UAS-ATX2-3F #4

Pdf-GAL4/+; UAS-ATX2-3F #6/+

Pdf-GAL4/ UAS-ATX2ΔPAM2-3F #6

Pdf-GAL4/+; UAS-ATX2ΔPAM2-3F #8/+

References and Notes

1. C. J. Doherty, S. A. Kay, Circadian control of global gene expression patterns. Annu. Rev. Genet. 44, 419 (2010). doi:10.1146/annurev-genet-102209-163432 Medline

2. X. Zheng, A. Sehgal, Speed control: cogs and gears that drive the circadian clock. Trends Neurosci. 35, 574 (2012). doi:10.1016/j.tins.2012.05.007 Medline

3. S. Kojima, D. L. Shingle, C. B. Green, Post-transcriptional control of circadian rhythms. J. Cell Sci. 124, 311 (2011). doi:10.1242/jcs.065771 Medline

4. D. Staiger, R. Green, RNA-based regulation in the plant circadian clock. Trends Plant Sci. 16, 517 (2011). doi:10.1016/j.tplants.2011.06.002 Medline

5. S. Bradley, S. Narayanan, M. Rosbash, NAT1/DAP5/p97 and atypical translational control in the Drosophila Circadian Oscillator. Genetics 192, 943 (2012). doi:10.1534/genetics.112.143248 Medline

6. C. Lim et al., The novel gene twenty-four defines a critical translational step in the Drosophila clock. Nature 470, 399 (2011). doi:10.1038/nature09728 Medline

7. G. Imbert et al., Cloning of the gene for spinocerebellar ataxia 2 reveals a locus with high sensitivity to expanded CAG/glutamine repeats. Nat. Genet. 14, 285 (1996). doi:10.1038/ng1196-285 Medline

8. S. M. Pulst et al., Moderate expansion of a normally biallelic trinucleotide repeat in spinocerebellar ataxia type 2. Nat. Genet. 14, 269 (1996). doi:10.1038/ng1196-269 Medline

9. K. Sanpei et al., Identification of the spinocerebellar ataxia type 2 gene using a direct identification of repeat expansion and cloning technique, DIRECT. Nat. Genet. 14, 277 (1996). doi:10.1038/ng1196-277 Medline

10. A. C. Elden et al., Ataxin-2 intermediate-length polyglutamine expansions are associated with increased risk for ALS. Nature 466, 1069 (2010). doi:10.1038/nature09320 Medline

11. O. A. Ross et al., Ataxin-2 repeat-length variation and neurodegeneration. Hum. Mol. Genet. 20, 3207 (2011). doi:10.1093/hmg/ddr227 Medline

12. D. A. Mangus, N. Amrani, A. Jacobson, Pbp1p, a factor interacting with Saccharomyces cerevisiae poly(A)-binding protein, regulates polyadenylation. Mol. Cell. Biol. 18, 7383 (1998). Medline

13. T. F. Satterfield, L. J. Pallanck, Ataxin-2 and its Drosophila homolog, ATX2, physically assemble with polyribosomes. Hum. Mol. Genet. 15, 2523 (2006). doi:10.1093/hmg/ddl173 Medline

14. R. Ciosk, M. DePalma, J. R. Priess, ATX-2, the C. elegans ortholog of ataxin 2, functions in translational regulation in the germline. Development 131, 4831 (2004). doi:10.1242/dev.01352 Medline

15. M. Kaneko, J. C. Hall, Neuroanatomy of cells expressing clock genes in Drosophila: transgenic manipulation of the period and timeless genes to mark the perikarya of

http://dx.doi.org/10.1146/annurev-genet-102209-163432http://dx.doi.org/10.1146/annurev-genet-102209-163432http://dx.doi.org/10.1016/j.tins.2012.05.007http://dx.doi.org/10.1016/j.tins.2012.05.007http://dx.doi.org/10.1242/jcs.065771http://dx.doi.org/10.1242/jcs.065771http://dx.doi.org/10.1016/j.tplants.2011.06.002http://dx.doi.org/10.1016/j.tplants.2011.06.002http://dx.doi.org/10.1534/genetics.112.143248http://dx.doi.org/10.1534/genetics.112.143248http://dx.doi.org/10.1038/nature09728http://dx.doi.org/10.1038/nature09728http://dx.doi.org/10.1038/ng1196-285http://dx.doi.org/10.1038/ng1196-285http://dx.doi.org/10.1038/ng1196-269http://dx.doi.org/10.1038/ng1196-269http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=8896555&dopt=Abstracthttp://dx.doi.org/10.1038/ng1196-277http://dx.doi.org/10.1038/ng1196-277http://dx.doi.org/10.1038/nature09320http://dx.doi.org/10.1038/nature09320http://dx.doi.org/10.1093/hmg/ddr227http://dx.doi.org/10.1093/hmg/ddr227http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=9819425&dopt=Abstracthttp://dx.doi.org/10.1093/hmg/ddl173http://dx.doi.org/10.1093/hmg/ddl173http://dx.doi.org/10.1242/dev.01352http://dx.doi.org/10.1242/dev.01352

circadian pacemaker neurons and their projections. J. Comp. Neurol. 422, 66 (2000). doi:10.1002/(SICI)1096-9861(20000619)422:13.0.CO;2-2 Medline

16. B. Grima, E. Chélot, R. Xia, F. Rouyer, Morning and evening peaks of activity rely on different clock neurons of the Drosophila brain. Nature 431, 869 (2004). doi:10.1038/nature02935 Medline

17. S. C. Renn, J. H. Park, M. Rosbash, J. C. Hall, P. H. Taghert, A pdf neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell 99, 791 (1999). doi:10.1016/S0092-8674(00)81676-1 Medline

18. D. Stoleru, Y. Peng, J. Agosto, M. Rosbash, Coupled oscillators control morning and evening locomotor behaviour of Drosophila. Nature 431, 862 (2004). doi:10.1038/nature02926 Medline

19. T. F. Satterfield, S. M. Jackson, L. J. Pallanck, A Drosophila homolog of the polyglutamine disease gene SCA2 is a dosage-sensitive regulator of actin filament formation. Genetics 162, 1687 (2002). Medline

20. I. Lastres-Becker et al., Insulin receptor and lipid metabolism pathology in ataxin-2 knock-out mice. Hum. Mol. Genet. 17, 1465 (2008). doi:10.1093/hmg/ddn035 Medline

21. C. McCann et al., The Ataxin-2 protein is required for microRNA function and synapse-specific long-term olfactory habituation. Proc. Natl. Acad. Sci. U.S.A. 108, E655 (2011). doi:10.1073/pnas.1107198108 Medline

22. C. Y. Chu, T. M. Rana, Translation repression in human cells by microRNA-induced gene silencing requires RCK/p54. PLoS Biol. 4, e210 (2006). doi:10.1371/journal.pbio.0040210 Medline

23. U. Nonhoff et al., Ataxin-2 interacts with the DEAD/H-box RNA helicase DDX6 and interferes with P-bodies and stress granules. Mol. Biol. Cell 18, 1385 (2007). doi:10.1091/mbc.E06-12-1120 Medline

24. A. Eulalio, E. Huntzinger, E. Izaurralde, GW182 interaction with Argonaute is essential for miRNA-mediated translational repression and mRNA decay. Nat. Struct. Mol. Biol. 15, 346 (2008). doi:10.1038/nsmb.1405 Medline

25. M. R. Fabian et al., Mammalian miRNA RISC recruits CAF1 and PABP to affect PABP-dependent deadenylation. Mol. Cell 35, 868 (2009). doi:10.1016/j.molcel.2009.08.004 Medline

26. I. Tuin et al., Stages of sleep pathology in spinocerebellar ataxia type 2 (SCA2). Neurology 67, 1966 (2006). doi:10.1212/01.wnl.0000247054.90322.14 Medline

27. L. Velázquez-Pérez et al., Sleep disorders in spinocerebellar ataxia type 2 patients. Neurodegener. Dis. 8, 447 (2011). doi:10.1159/000324374 Medline

28. D. J. Dijk, C. A. Czeisler, Paradoxical timing of the circadian rhythm of sleep propensity serves to consolidate sleep and wakefulness in humans. Neurosci. Lett. 166, 63 (1994). doi:10.1016/0304-3940(94)90841-9 Medline

http://dx.doi.org/10.1002/(SICI)1096-9861(20000619)422:1%3c66::AID-CNE5%3e3.0.CO;2-2http://dx.doi.org/10.1002/(SICI)1096-9861(20000619)422:1%3c66::AID-CNE5%3e3.0.CO;2-2http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=10842219&dopt=Abstracthttp://dx.doi.org/10.1038/nature02935http://dx.doi.org/10.1038/nature02935http://dx.doi.org/10.1016/S0092-8674(00)81676-1http://dx.doi.org/10.1016/S0092-8674(00)81676-1http://dx.doi.org/10.1038/nature02926http://dx.doi.org/10.1038/nature02926http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=12524342&dopt=Abstracthttp://dx.doi.org/10.1093/hmg/ddn035http://dx.doi.org/10.1093/hmg/ddn035http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=18250099&dopt=Abstracthttp://dx.doi.org/10.1073/pnas.1107198108http://dx.doi.org/10.1073/pnas.1107198108http://dx.doi.org/10.1371/journal.pbio.0040210http://dx.doi.org/10.1371/journal.pbio.0040210http://dx.doi.org/10.1091/mbc.E06-12-1120http://dx.doi.org/10.1091/mbc.E06-12-1120http://dx.doi.org/10.1038/nsmb.1405http://dx.doi.org/10.1038/nsmb.1405http://dx.doi.org/10.1016/j.molcel.2009.08.004http://dx.doi.org/10.1016/j.molcel.2009.08.004http://dx.doi.org/10.1212/01.wnl.0000247054.90322.14http://dx.doi.org/10.1212/01.wnl.0000247054.90322.14http://dx.doi.org/10.1159/000324374http://dx.doi.org/10.1159/000324374http://dx.doi.org/10.1016/0304-3940(94)90841-9http://dx.doi.org/10.1016/0304-3940(94)90841-9

ATAXIN-2 Activates PERIOD Translation to Sustain Circadian Rhythms in DrosophilaReferences and Notes