TRANSLATIONAL CONTROL

Fragile X mental retardation 1 geneenhances the translation of largeautism-related proteinsEthan J. Greenblatt and Allan C. Spradling*

Mutations in the fragile X mental retardation 1 gene (FMR1) cause the most commoninherited human autism spectrum disorder. FMR1 influences messenger RNA (mRNA)translation, but identifying functional targets has been difficult. We analyzed quiescentDrosophila oocytes, which, like neural synapses, depend heavily on translatingstored mRNA. Ribosome profiling revealed that FMR1 enhances rather than repressesthe translation of mRNAs that overlap previously identified FMR1 targets, and actspreferentially on large proteins. Human homologs of at least 20 targets are associatedwith dominant intellectual disability, and 30 others with recessive neurodevelopmentaldysfunction. Stored oocytes lacking FMR1 usually generate embryos with severeneural defects, unlike stored wild-type oocytes, which suggests that translation of multiplelarge proteins by stored mRNAs is defective in fragile X syndrome and possibly otherautism spectrum disorders.

FMR1 is a polysome-associated RNA bindingprotein required for the nervous systemand ovary to develop and function normallyin humans,mice, andDrosophila (1, 2). Bothtissues translationally control storedmRNAs

associated with FMR1-containing ribonucleo-protein particles (RNPs) (3–6), which suggeststhat FMR1 has a specific function in using storedmRNAs. However, the challenge of obtaininghighly enriched FMR1-containing RNPs fromneural tissuemay have contributed to difficultiesin defining FMR1 target genes (7–9). We reasonedthat FMR1 function could be studied in a physio-logically relevant context usingmatureDrosophilaoocytes, which lack transcription and dependentirely on ongoing translation.Drosophila oogenesis is highly amenable to

such studies because each female can hold upto 30 mature oocytes per ovary for several weeks(Fig. 1A). In the ovary, each oocyte is surroundedby 800 somatic cells constituting a follicle. Wefound a reliablemethod (see supplementarymate-rials) to maintain completed follicles in the ovaryfor a known period of time in the absence of newfollicle maturation (Fig. 1B). Ovulation and fer-tilization could then be induced by adding malesto test the stored oocyte’s ability to supportembryogenesis.We tested the function of specific genes during

oocyte storage by depleting their transcripts withgermline-specific GAL4-driven RNA interference(RNAi), which is produced throughout oogenesisstarting in the germline stem cell. Disrupting agene required during oocyte storage would causeoocytes to at first develop normally, and then tolose developmental capacity more rapidly thanthe wild type during further storage. To screen

for such genes, we depleted Fmr1 mRNA or sev-eral other candidate gene transcripts and analyzedoocytes after a storage period of 1 day or 10 days(Fig. 1C). In most gene knockdown lines and inwild-type controls, the hatch rate was nearly100% after 1 day of storage, and after 10 daysonly dropped to 80%. In contrast, Fmr1-depletedoocytes hatched normally at first, but after 10 daysof storage, only 20 to 30% of them completeddevelopment (Fig. 1C). Germline Fmr1 RNAi dras-

tically reduced Fmr1 mRNA levels, and antibodystaining confirmed that FMR1 protein was effec-tively depleted throughout oogenesis specifically ingerm cells but not in somatic cells (fig. S1).We validated the Fmr1 requirement in stored

oocytes by showing that oocyte viability dropscontinuously over time without Fmr1 (Fig. 1D).The defects were specific to the old oocytes them-selves. Refeeding the mother caused the remain-ing stored oocytes to be laid and new oocytes tomature, which were fully functional (Fig. 1D).Thus, germline Fmr1 RNAi does not observablyimpair stem cells, follicle development, or anynonautonomous aspects of female germline func-tion. This differs from Fmr1 mutants, which losestem cells and produce fewer follicles becausemutants affect niche cells and other somatic cellsin addition to germ cells (10).We analyzed embryos derived from control

and FMR1-deficient oocytes to investigate whatprocesses are affected by FMR1 depletion. Em-bryos from control oocytes developed a normalnervous system regardless of prior storage, asshown by staining with broadly expressed neuralmarkers (Fig. 2, A and B). The same was true ofembryos derived from Fmr1 RNAi oocytes after1 day of storage (Fig. 2A). However, in embryosderived from follicles lacking FMR1 during 10 daysof storage, ventral nerve cord–specific labelingshowed missing commissures and breaks in thelongitudinal connectives (Fig. 2C), in contrast tothe wild type. These neuronal defects were notdue to a generalized deterioration of the FMR1-depleted oocyte’s ability to support embryonicdevelopment (fig. S2). We never observed com-parable neural defects in embryos derived from

RESEARCH

Greenblatt et al., Science 361, 709–712 (2018) 17 August 2018 1 of 4

Howard Hughes Medical Institute Research Laboratories,Department of Embryology, Carnegie Institution for Science,Baltimore, MD 21218, USA.*Corresponding author. Email: spradling@ciwemb.edu

D

A

% H

atch

ed

Oocyte Age (Days)

n.s.n.s.

p=0.047n.s.

p=0.026

p=0.038

p=0.006

p=0.003

p=0.037n.s.

0

25

50

75

100

1 4 7 10

GF

P C

on

tro

lF

mr1

RN

Ai

#1

Fm

r1 R

NA

i #2

FollicleCells

GermlineStem Cells

NurseCells

Arrested Mature FolliclesContinued Growth

OocyteFollicle

Cell Layer

Stage 14 Stage 14

C

0

50

100

Ore

R C

ontr

ol

Yps

#1

Yps

#2

Fm

r1 #

1

Fm

r1 #

2

FO

XO

#1

FO

XO

#2

Me3

1B

Rox

8

MK

2 #

1

MK

2 #

2

Pu

m #

1

Pu

m #

2% H

atch

ed

1 Day 10 DaysOocyte age:Germline RNAi

B

0

10

20

30

40

0 1 2 3 4 5 6 7 8 9 1011121314# F

ollic

les

/ Ova

ry

Day

Arrested Mature Follicles

Post-checkpointDeveloping Follicles

+ Yeast – Yeast

NutrientCheckpoint

Stage 8

RefedControl

Fig. 1. Fmr1 is specifically required during the storage of mature, quiescent stage 14 oocytesin the ovary. (A) Schematic of a Drosophila ovariole with immature precheckpoint follicles and twostored mature stage 14 follicles. (B) Arrested mature follicle stability (red) following feeding protocolas described (see supplementary materials). (C) Fmr1 knockdown (lines 1 and 2), but notknockdowns of controls or other indicated genes, specifically reduces 10-day stored (red) but not1-day stored (black) oocytes from developing into hatching larvae. (D) Fmr1 germline RNAiduring storage progressively reduces the fraction of mature oocytes competent after 1, 4, 7, or10 days of storage to support development. Refeeding females to promote maturation of fresh stage8 follicles restores full developmental potential. Error bars in (B) and (D) denote SD; n.s., notsignificant. GFP, green fluorescent protein.

on May 18, 2021

http://science.sciencem

ag.org/D

ownloaded from

wild-type oocytes, even when they were storedfor 14 days when fewer than 20% of their em-bryos developed to hatching (Fig. 2D). In contrast,more than 50% of embryos derived from FMR1-deficient oocytes stored in the ovary for 10 daysdeveloped a severely abnormal nervous system(Fig. 2E). Thus, disrupting FMR1 function whileoocytes are fully dependent on translational regu-lation specifically compromised their ability tosupport neural development relative to controls.To determine how FMR1 disruption affects

oocyte translation and to identify FMR1 trans-lational target genes that may be important forsustaining nervous system development, we de-veloped a ribosome-profiling protocol based on(11) (see supplementary materials) to quantifyoocyte translation in an unbiased, genome-widemanner. Flies induced for oocyte storage wereanalyzed after only 1 or 2 days to identify initialtranslational changes in healthy Fmr1 knockdownoocytes prior to viability reduction. Althoughwhole ovaries were needed to obtain enoughmaterial for profiling, most of the analyzedribosomes should still derive from stored stage14 follicles, because they are larger than the totalsize of earlier follicles. Interestingly, the ribosomefootprints and transcript levels of most mRNAswere unaffected by germline Fmr1 knockdown(Fig. 3A); this result indicates that FMR1 does notgenerally control translation or mRNA stability.From 11 independent, highly reproducible ribo-

some profiling experiments (fig. S3), we identified421 genes whose germline expression signifi-cantly declined and 14 genes that significantlyincreased expression in Fmr1 RNAi (Fig. 3, Band C, and tables S1 and S2). Except for Fmr1itself, translation of the significantly altered tar-gets generally decreased by a factor of 1.3 to 2.5,which we verified by Western blotting (fig. S4),whereas theirRNA levelswereunchanged (Fig. 3B).Many down-regulated genes, at least 56 of 421,are orthologs of human genes that have been im-plicated in human neurodevelopmental syn-dromes (Fig. 3C and table S3)—a proportiongreater than expected by chance (P = 1.1 × 10–9;fig. S5A). For example, the neurofibromatosisgene Nf1, which is associated with cognitive andbehavioral disorders and neural tumors (12), wasreduced by a factor of 2.5. Several E3 ubiquitinligases, including CTRIP/TRIP12, POE/UBR4, andHUWE1, whose human homologs are associatedwith intellectual disability, autism, early-onset de-mentia, and schizophrenia (13–15), were reducedby about a factor of 2. In total, homologs of atleast 20 dominant autism/intellectual disabilitygenes were significantly reduced (table S3). Be-cause mutations in these genes are dominant(16), a factor of 2 reduction in expression has po-tential consequences, even for a single target.To determine whether FMR1 acted on target

transcripts through direct binding, we comparedour candidate FMR1 targetswith aprevious report,which used proximity-based strategies to cross-link mRNAs in brain tissue before immuno-precipitation of FMR1 (7). We found significantoverlap between both datasets (P = 1.2 × 10–25;fig. S5B), which suggests that FMR1 directly

binds many affected transcripts. However, ourresults differed from prior studies in two impor-tant ways. First, the great majority of targetsdecreased in expression, indicating that FMR1usually enhances rather than represses transla-tion, in contrast to most (17, 18) but not all (9, 19)previous reports. This difference might arise be-cause multiple FMR1 targets act negatively onprotein stability, translation, or cell growth, in-cluding nine ubiquitin ligases, Nf1, and Not1(table S1). Down-regulation of these negativeregulators after Fmr1 loss might substantiallyincrease protein levels, simulating the direct ef-fects of a translational repressor.Second, almost all of the affected proteins are

much larger in size than the average Drosophilaprotein. Dividing mRNAs undergoing transla-tion into size classes showed that FMR1 strik-ingly affects translation in proportion to proteinsize (Fig. 3, C and D) and to some extent UTRlength (fig. S6A), but not by transcript level(fig. S6D). The translation of nearly half (46%)of expressed proteins longer than 2000 aminoacids, 13% of proteins 750 to 1000 amino acidsin length, but only 1% of proteins shorter than250 amino acids was significantly reduced withFmr1 knockdown (Fig. 3C). Fmr1 RNAi did notalways impair the translation of large proteins,as therewas a broad response toFmr1 knockdown(Fig. 3C) and the translation of many large pro-teins was only minimally affected. FMR1 targetshad low translation efficiencies (TEs) in oocytes(Fig. 3D), similar to long transcripts generally(fig. S6B). However, FMR1 boosted the translationof affected long transcripts regardless of TE(Fig. 3E), implicating size and not low TE as thepredominant factor. The size effect was not dueto reduced processivity, because we observed auniform reduction in footprints across the entirecoding sequence of target mRNAs (Fig. 3F). The

preferential effect on large mRNAs is likely me-diated by direct FMR1 binding, because theaverage size of target proteins common to boththis study and (7) was 1841 amino acids. Further-more, genes linked to autism as a group are sig-nificantly longer than average (16, 20).The Poe/Ubr4 gene, encoding one of the lon-

gestDrosophila proteins (5322 amino acids), wasinvestigated as an FMR1 target with potentiallylarge effects, something not previously identified.Both Poe and Fmr1 mutant Drosophila are malesterile, cannot fly, and show increased neuro-muscular junction synaptic excitability (21–23).Stored Poe mutant oocytes lost developmentalcompetence at the same rate as Fmr1 germlineRNAi oocytes (Fig. 4A), and these embryos alsodeveloped a high frequency of neural defects(Fig. 4B and fig. S7A). Beginning in nearly ma-ture follicles, POE protein formed 0.5- to 2-mmspherical particles in germ cells (Fig. 4C), whichwere distinct from RNP granules containing theP body marker TRAL (fig. S8). POE antibodystaining was lost in Poe germline RNAi and Poemutants (Fig. 4D). As predicted, POE proteinlevels were reduced and particles were reduced oreliminated in Fmr1 germline RNAi and in Fmr1-null egg chambers (Fig. 4D). These observationsdemonstrate that Poe is a major functionaltarget of FMR1, and that POE is itself essentialto maintain the oocyte’s ability to support neuraldevelopment. However, overexpression of POEusing a duplication in an Fmr1RNAi backgroundrestored POE granule expression (Fig. 4D) butdid not rescue Fmr1RNAi lethality upon storage(fig. S7F).FMR1’s function in maintaining an oocyte’s

ability to support neural development may onlybe revealed during oocyte storage because someFMR1 targets act catalytically. For example, E3ligaseswhose levels are reduced in FMR1-deficient

Greenblatt et al., Science 361, 709–712 (2018) 17 August 2018 2 of 4

Fig. 2. Stored FMR1-depletedoocytes frequently generateembryos with neural defects.(A) Control oocytes stored in vivofor 1 day or 10 days support normalembryonic nervous system devel-opment (22C10 antibody). Fmr1RNAi oocytes stored in vivo for1 day support normal developmentbut after storage for 10 days pro-duce highly abnormal nervoussystem development. (B) Normalventral nerve cord (BP102 anti-body) from embryo developed from10-day control oocyte. (C) Abnor-mal ventral nerve cords includingbroken or fused connectives(arrows) from three embryosdeveloped from 10-day Fmr1 RNAioocytes. (D) Normal nerve cordfrom a control oocyte stored for14 days. (E) Summary of nervoussystem development in embryosfrom control (GFP) and Fmr1 RNAioocytes (N ≥ 50). All scale bars, 20 µm.

A B

C

D

0

50

100

1

Control

10 1

Fmr1 RNAi

10

Normal%

Em

bry

os

NeuralDefect

E

Control

Ooc

yte

Age

(D

ays)

Oocyte Age (Days)

Fmr1 RNAi

1

10

RESEARCH | REPORTon M

ay 18, 2021

http://science.sciencemag.org/

Dow

nloaded from

oocytes might no longer be able to prevent theoveraccumulation of some of their target proteins,which might eventually reach levels in the storedoocytes that interfere with neural development.FMR1 regulates translation both in the ovary

and during neural development as part of RNPgranules that may either stimulate or represstranslation. Studies of these granules suggest a

potential explanation for the size effect we ob-served on target proteins. Mammalian and yeastmRNAs that are large and/or inefficiently trans-lated preferentially associate with stress granuleswhere their translation is repressed (24). FMR1may function on specific targets to counteract theinherent general tendency of large mRNAs to besegregated into inactive RNP particles. Alterna-

tively, FMR1 might directly or indirectly promotetranslation initiation in association with RNPs,or it might affect mRNA transport along micro-tubules to sites of active initiation.We propose that maintaining the translation

of largemRNAs in the translational environmentsexperienced far from the nucleus at synapses inmany neurons represents a general challenge that

Greenblatt et al., Science 361, 709–712 (2018) 17 August 2018 3 of 4

D

C

6

4

2

0-2 0 2

6

4

2

0-2 0 2

6

4

2

0-2 0 2

46 0

13 0

1 2

7

6

5

4

3

2

1

0

>2000>1000

>750>500

>250

<250

# Amino Acids

-2 -1 0 1 2

421 14Ctrip

Nf1

Poe

ProSAPHuwe1

OsaBrwd3

-lo

g 10 (P

-va

lue

)

Fold-change in Translationlog2 Fmr1 RNAi / Control

Control RNAiFmr1 RNAi

Nucleotide Position (kB)

Poe Top2

Nor

mal

ized

Rea

d D

epth

0

0.8

1.6

0

0.8

1.2

0.4

0 4 0 1 2 3 4 58 12 16

Huwe1

0

0.8

1.2

0.4

0 4 8 12 16

Orb

0

0.8

1.2

0.4

0 1 2 3 4

0

0.5

1.0

-1 0 1

Cum

ulat

ive

Freq

uenc

y >2000>1000>750>500>250<250

# Amino Acids

0

0.5

1.0

-1 0 1

>21–2.5–1.25–.5.15–.25<.15

TE

Fold-change in Translationlog2 Fmr1 RNAi / Control

E

Cum

ulat

ive

Freq

uenc

y

Fmr1RNAi

ControlRNAi

# Amino Acids:> 2000 1000–2000

log2 TE

All Transcripts

0

0.5

1.0

-3 -2 -1 0 10

0.5

1.0

-3 -2 -1 0 10

0.5

1.0

-3 -2 -1 0 1

mRNA-seq

ARibosome

FootprintingR2 = 0.99

0 1 2 3 4 5012345

0 1 2 3 4 5012345

R2 = 0.95

Fm

r1 R

NA

ilo

g 10 T

PM

Fm

r1 R

NA

ilo

g 10 T

PM

Control RNAilog10 TPM

Control RNAilog10 TPM

B

F

Sbf

mRNA-seqRibosome Footprinting0 0.5 1 1.5

Fmr1Nf1

Ana3Huwe1

FryPoe

CtripIntS1

CG15099CG14967

Nup188Rme-8

CG16974Shtd

ProSAPFaf

Epg5Ulp1

GramD1BNot1

6822765

1978

5147

3545

5323

3141

2054

2600

2301

1867

2408

1258

2031

1872

2762

2456

1514

1207

2504

# A

min

o A

cids

Gen

e

TPM Fold ChangeFmr1 RNAi/Control:

(%)

Fig. 3. FMR1 stimulates translation during storage of transcripts frommultiple intellectual disability and autism genes. (A) Translationalprofile [log10(TPM), transcripts per million] and mRNA abundance profile[mRNA-seq, log10(TPM)] are highly similar between control and Fmr1 RNAioocytes (stored 1 to 2 days). (B) Top genes translationally reduced byFmr1 RNAi from 11 ribosome-footprinting experiments do not showsignificant changes in mRNA levels. (C) Significance versus relative (fold)change plot reveals 421 candidate targets translationally stimulated by

FMR1 (P < 0.01, t test). Protein size class is indicated by color.(D) Cumulative plot of translation (Fmr1 RNAi/control) as a function ofprotein size (left) or translational efficiency (TE, right) defined asribosome-footprinting TPM(Fmr1 RNAi)/mRNA-seq TPM. (E) Translationof large mRNAs in Fmr1 RNAi versus controls is reduced independentof TE. (F) Normalized read depth is plotted for two FMR1 targets (Poe andHuwe1) and two nontargets (Orb and Top2). In Fmr1 RNAi oocytes,target gene footprints are reduced at all positions along the mRNA.

D

Stage 8

Stage 8

Stage 10B(End of follicle growth)

Stage 10B

WT Poe01659

Fmr13

Fmr1Δ50Fmr1 RNAiFmr1 RNAiPoe OE

Poe RNAi

0

50

100

WTDay 1

PoeDay 1

WTDay 8

PoeDay 8

Normal

NeuralDefect

% E

mb

ryo

s

C

B

A OreR / ywmCherry RNAiPoe01659 / +

Fmr1 RNAi

Poe01659

Poe03420

0

25

50

75

100

1 4 7 10

Ha

tch

Ra

te (%

)

Oocyte Age (Days)

Fig. 4. Poe is required for oocyte storage and neural development.(A) Poe mutation accelerates oocyte decline during storage. Error barsdenote SD. (B) Poe mutant oocytes frequently fail to support normal neuraldevelopment after prolonged storage. (C) POE antibody staining(see supplementary materials) during follicle development, showing

germline granules that arise in maturing follicles. Scale bars, 3 mm.(D) Many POE granules are seen in wild-type stage 10 follicles but notin Poe RNAi, Poe01659, Fmr1 RNAi, or null Fmr13/D50 follicles. Fmr1 RNAicombined with POE overexpression (OE) recovers POE granules.Scale bars, 20 mm.

RESEARCH | REPORTon M

ay 18, 2021

http://science.sciencemag.org/

Dow

nloaded from

not only underlies fragile X syndrome but isrelevant to other neurodevelopmental condi-tions. The same challenges likely exist in oocytes,spermatocytes, and non-neural somatic cells thatrequire regulated translation from storedmRNAs.Because the challenges of translating large pro-teins are likely to increase in adult neurons underthe influence of aging, the pathways and targetsassayed here may contribute to adult-onset neuralimpairments such as schizophrenia and dementia.Improved knowledge of how FMR1 preservestarget translation, and of the identities of majortarget genes such as Poe/Ubr4, may open newopportunities to monitor susceptible cells and tointervene to mitigate declining levels of the mostcritical targets. Small-molecule agents that coun-teract the tendency of large mRNAs to be segre-gated into inactive granules represent potentiallyvaluable therapeutics. Continued study of thesehighly conserved pathways in Drosophila repre-sents one powerful and efficient means to furtheraddress both the fundamental and applied im-plications of these findings.

REFERENCES AND NOTES

1. R. J. Hagerman et al., Nat. Rev. Dis. Primers 3, 17065 (2017).2. M. Drozd, B. Bardoni, M. Capovilla, Front. Mol. Neurosci. 11, 124

(2018).3. S. A. Barbee et al., Neuron 52, 997–1009 (2006).4. R. Rosario et al., PLOS ONE 11, e0163987 (2016).5. S. B. Christie, M. R. Akins, J. E. Schwob, J. R. Fallon,

J. Neurosci. 29, 1514–1524 (2009).6. A. Costa et al., Dev. Cell 8, 331–342 (2005).7. J. C. Darnell et al., Cell 146, 247–261 (2011).8. M. Ascano Jr. et al., Nature 492, 382–386 (2012).9. R. Tabet et al., Proc. Natl. Acad. Sci. U.S.A. 113, E3619–E3628 (2016).10. L. Yang et al., Hum. Mol. Genet. 16, 1814–1820 (2007).11. J. G. Dunn, C. K. Foo, N. G. Belletier, E. R. Gavis,

J. S. Weissman, eLife 2, e01179 (2013).12. S. M. Morris et al., JAMA Psychiatry 73, 1276–1284 (2016).13. J. Zhang et al., Hum. Genet. 136, 377–386 (2017).14. D. Monies et al., Hum. Genet. 136, 921–939 (2017).15. M. Bosshard et al., Sci. Rep. 7, 15050 (2017).16. W. Pereanu et al., Nucleic Acids Res. 46, D1049–D1054 (2018).17. F. Zalfa et al., Cell 112, 317–327 (2003).18. B. Laggerbauer, D. Ostareck, E. M. Keidel, A. Ostareck-Lederer,

U. Fischer, Hum. Mol. Genet. 10, 329–338 (2001).19. E. G. Bechara et al., PLOS Biol. 7, e1000016 (2009).20. I. F. King et al., Nature 501, 58–62 (2013).21. S. Richards, T. Hillman, M. Stern, Genetics 142, 1215–1223 (1996).22. J. J. Fabrizio, G. Hime, S. K. Lemmon, C. Bazinet, Development

125, 1833–1843 (1998).

23. Y. Q. Zhang et al., Cell 107, 591–603 (2001).24. A. Khong et al., Mol. Cell 68, 808–820.e5 (2017).

ACKNOWLEDGMENTS

We thank A. Pinder for help generating RNA-seq libraries, J. Dunn(UCSF) for technical advice for our ribosome footprintingexperiments, and E. Wagner (UTMB) and T. Jongens (UPenn) forgenerously providing the INTS1 antibody and the Fmr13 mutantfly strain, respectively. Funding: Supported by the Jane CoffinChilds Memorial Fund (E.G.) and the Howard Hughes MedicalInstitute (A.C.S.). Author contributions: Both E.G. andA.C.S. contributed to all aspects of this research except that E.G.performed the research. Competing interests: The authorsdeclare no competing interests. Data availability: Ribosomeprofiling and mRNA-seq data reported here are available fromthe NCBI BioProject website under accession numberPRJNA466150.

SUPPLEMENTARY MATERIALS

www.sciencemag.org/content/361/6403/709/suppl/DC1Materials and MethodsFigs. S1 to S8Tables S1 to S5References (25–27)

13 January 2018; resubmitted 5 May 2018Accepted 27 June 201810.1126/science.aas9963

Greenblatt et al., Science 361, 709–712 (2018) 17 August 2018 4 of 4

RESEARCH | REPORTon M

ay 18, 2021

http://science.sciencemag.org/

Dow

nloaded from

Fragile X mental retardation 1 gene enhances the translation of large autism-related proteinsEthan J. Greenblatt and Allan C. Spradling

DOI: 10.1126/science.aas9963 (6403), 709-712.361Science 

, this issue p. 709; see also p. 648Scienceinactive ribonucleoprotein particles. This mechanism may underlie other causes of autism and mental dysfunction.many associated with autism. FMR1 seems to help maintain translation of large mRNAs that otherwise condense intoRibosome profiling of oocytes identified a specific role for FMR1 in enhancing the translation of large proteins, including

loss leads to oocytes that generate embryos exhibiting neural defects (see the Perspective by Aryal and Klann).Fmr1 oocytes, Greenblatt and Spradling found thatDrosophilathought to reduce protein synthesis (translation) at synapses. In

isprimary ovarian insufficiency, which are prominent intellectual disability and reproductive disorders, respectively. FMR1 associated−) gene underlie fragile X syndrome and fragile XFMR1Mutations in the fragile X mental retardation 1 (

Fragile X and fragile translation in flies

ARTICLE TOOLS http://science.sciencemag.org/content/361/6403/709

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2018/08/15/361.6403.709.DC1

CONTENTRELATED

http://stm.sciencemag.org/content/scitransmed/10/452/eaar4338.fullhttp://stm.sciencemag.org/content/scitransmed/8/321/321ra5.fullhttp://stm.sciencemag.org/content/scitransmed/8/336/336ra61.fullhttp://science.sciencemag.org/content/sci/361/6403/648.full

REFERENCES

http://science.sciencemag.org/content/361/6403/709#BIBLThis article cites 27 articles, 4 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2018 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on May 18, 2021

http://science.sciencem

ag.org/D

ownloaded from