Neural circuit architecture defects in a Drosophila of Fragile X … · 2011. 6. 10. · neuromuscular junction (NMJ), clock neurons in the circadian activity circuit and Kenyon cells

INTRODUCTIONFragile X syndrome (FXS), the most common genetic determinantof cognitive impairment and autism spectrum disorders (Koukouiand Chaudhuri, 2007; Penagarikano et al., 2007), is caused solelyby the loss of the fragile X mental retardation 1 (FMR1) geneproduct (FMRP) (Pieretti et al., 1991). FMRP is an mRNA-bindingprotein known to regulate mRNA stability, mRNA trafficking andthe translation of a number of neuronal transcripts (Laggerbaueret al., 2001; Li et al., 2001; Lu et al., 2004; Muddashetty et al., 2007;Tessier and Broadie, 2008; Zhang et al., 2001). Clinically, FXS is awide-spectrum disorder, with patients displaying hyperactivity,disrupted sleep patterns and mild to severe intellectual disability,among other behavioral impairments (Einfeld et al., 1991; Elia et

al., 2000; Hagerman et al., 2010; Hagerman et al., 2009; Levenga etal., 2010; Penagarikano et al., 2007). At a cellular level, defects incortical dendritic spine morphology have been observed in FXSpatient brain autopsies (Hinton et al., 1991; Rudelli et al., 1985),suggesting immature synaptic connections. A great deal ofinvestigation in FXS disease models supports the conclusion thatFMRP plays a predominant role in the activity-dependentregulation of synaptic development and plasticity (Antar andBassell, 2003; Auerbach and Bear, 2010; Costa-Mattioli et al., 2009;Huber et al., 2002; Pan et al., 2008; Tessier and Broadie, 2008; Tessierand Broadie, 2010; Waung and Huber, 2009; Zhang and Broadie,2005). Given the high prevalence of this devastating neurologicalcondition, pharmacological treatments for FXS have long beensought.

Many studies support ‘the metabotropic glutamate receptor(mGluR) theory of FXS’, which suggests that enhanced mGluRsignaling causes defects in synaptogenesis, dendritic spinematuration, and long-term depression (LTD) and potentiation(LTP) (Antar et al., 2004; Antar and Bassell, 2003; Auerbach andBear, 2010; Bear, 2005; Bear et al., 2008; Bear et al., 2004; Dolenand Bear, 2008; Dolen et al., 2010; Dolen et al., 2007; Huber etal., 2002; Meredith et al., 2010; Pan et al., 2008; Pan et al., 2004;Penagarikano et al., 2007; Repicky and Broadie, 2009; Waung andHuber, 2009). These findings have made mGluRs the primary

Disease Models & Mechanisms 673

Disease Models & Mechanisms 4, 673-685 (2011) doi:10.1242/dmm.008045

1Department of Biological Sciences and Department of Cell and DevelopmentalBiology, Kennedy Center for Research on Human Development, VanderbiltUniversity, Nashville, TN 37232, USA*Author for correspondence ([email protected])

Received 16 November 2010; Accepted 3 May 2011

© 2011. Published by The Company of Biologists LtdThis is an Open Access article distributed under the terms of the Creative Commons AttributionNon-Commercial Share Alike License (http://creativecommons.org/licenses/by-nc-sa/3.0), whichpermits unrestricted non-commercial use, distribution and reproduction in any medium providedthat the original work is properly cited and all further distributions of the work or adaptation aresubject to the same Creative Commons License terms.

SUMMARY

Fragile X syndrome (FXS), caused by loss of the fragile X mental retardation 1 (FMR1) product (FMRP), is the most common cause of inherited intellectualdisability and autism spectrum disorders. FXS patients suffer multiple behavioral symptoms, including hyperactivity, disrupted circadian cycles, andlearning and memory deficits. Recently, a study in the mouse FXS model showed that the tetracycline derivative minocycline effectively remediatesthe disease state via a proposed matrix metalloproteinase (MMP) inhibition mechanism. Here, we use the well-characterized Drosophila FXS modelto assess the effects of minocycline treatment on multiple neural circuit morphological defects and to investigate the MMP hypothesis. We first treatDrosophila Fmr1 (dfmr1) null animals with minocycline to assay the effects on mutant synaptic architecture in three disparate locations: theneuromuscular junction (NMJ), clock neurons in the circadian activity circuit and Kenyon cells in the mushroom body learning and memory center.We find that minocycline effectively restores normal synaptic structure in all three circuits, promising therapeutic potential for FXS treatment. Wenext tested the MMP hypothesis by assaying the effects of overexpressing the sole Drosophila tissue inhibitor of MMP (TIMP) in dfmr1 null mutants.We find that TIMP overexpression effectively prevents defects in the NMJ synaptic architecture in dfmr1 mutants. Moreover, co-removal of dfmr1similarly rescues TIMP overexpression phenotypes, including cellular tracheal defects and lethality. To further test the MMP hypothesis, we generateddfmr1;mmp1 double null mutants. Null mmp1 mutants are 100% lethal and display cellular tracheal defects, but co-removal of dfmr1 allows adultviability and prevents tracheal defects. Conversely, co-removal of mmp1 ameliorates the NMJ synaptic architecture defects in dfmr1 null mutants,despite the lack of detectable difference in MMP1 expression or gelatinase activity between the single dfmr1 mutants and controls. These resultssupport minocycline as a promising potential FXS treatment and suggest that it might act via MMP inhibition. We conclude that FMRP and TIMPpathways interact in a reciprocal, bidirectional manner.

Neural circuit architecture defects in a Drosophila modelof Fragile X syndrome are alleviated by minocyclinetreatment and genetic removal of matrixmetalloproteinaseSaul S. Siller1 and Kendal Broadie1,*

RESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

target of FXS drug discovery (Levenga et al., 2010). Specifically,mGluR inhibitors, such as 2-methyl-6-phenylethynyl-pyridine(MPEP), have proven to be effective at rescuing FXS defects inboth Drosophila and mouse models of the disease (Bolduc et al.,2008; Choi et al., 2010; Dolen and Bear, 2008; Dolen et al., 2010;McBride et al., 2005; Pan et al., 2008). In Drosophila, MPEPrescues cognitive impairments, synaptic plasticity, courtship andmushroom body learning defects in Drosophila Fmr1 (dfmr1) nullanimals. Genetic studies of mGluR mutants strongly support themechanism (Bear, 2005; Bear et al., 2008; Bear et al., 2004; Dolenand Bear, 2008; Dolen et al., 2010; Dolen et al., 2007; McBride etal., 2005; Pan and Broadie, 2007; Pan et al., 2008; Repicky andBroadie, 2009). Although MPEP cannot be used clinically owingto its toxicity, next generation mGluR antagonists are in clinicaltrials (Levenga et al., 2010; Wang et al., 2010). Lithium, aninhibitor of GSK3, which is a downstream effector of mGluRsignaling, has also been taken into FXS clinical trials and shownto have beneficial effects on the disease state (Berry-Kravis et al.,2008). Thus, the mGluR pathway is certainly one promisingavenue for FXS treatment.

In addition to mGluR inhibitors, a new possible FXS drugtreatment is the tetracycline derivative minocycline, the potentiallybeneficial effects of which were recently revealed in a studyinvolving a mouse model of FXS (Bilousova et al., 2009).Minocycline has previously been proven to be effective in treatinga surprising range of neurological disorders, including multiplesclerosis, Huntington’s disease, Parkinson’s disease and Alzheimer’sdisease (Choi et al., 2007; Kim and Suh, 2009; Popovic et al., 2002;Wang et al., 2003; Wu et al., 2002). In the mouse FXS model,minocycline promoted the maturation of hippocampal dendriticspines towards normal morphology, both in vitro and in vivo, andrepressed anxiety and memory defects in an elevated plus-mazetest (Bilousova et al., 2009). More recently, an open-label add-onminocycline trial of FXS patients reported a beneficial effect onfive out of six measured behaviors and improved the score that wasobtained on the ABC-C irritability scale (Paribello et al., 2010).Minocycline has several known targets, including, but not limitedto, p38 MAPK, iNOS and caspases (Kim and Suh, 2009). However,the mouse FXS study suggested that minocycline functionsspecifically by inhibiting matrix metalloproteinase-9 (MMP9)(Bilousova et al., 2009). The 24 mammalian MMPs are a family ofzinc-dependent extracellular proteases, which cleave membrane-associated and secreted proteins to remodel the extracellularmatrix (Ethell and Ethell, 2007; Page-McCaw et al., 2007; Rivera etal., 2010). MMPs are involved in normal neuronal development,function and plasticity, and are also implicated in severalneurological pathologies (Rivera et al., 2010). In FMR1 knockout(KO) mouse hippocampus, MMP9 expression and activity wereincreased, and minocycline decreased this enhancement (Bilousovaet al., 2009). MMP9 treatment of wild-type hippocampal neuronsin culture produced dendritic spine profiles that were similar toFMR1 KO neurons. Other MMPs (i.e. MMP2) were not detectablyinvolved. Four endogenous tissue inhibitors of matrixmetalloproteinases (TIMPs) are known in mammals (Stetler-Stevenson, 2008).

In this study, we use the Drosophila FXS model to test the effectsof minocycline on a broad range of neuronal architecturalphenotypes and investigate the hypothesized MMP mechanism.

We assay three neural cell types in dfmr1 null mutants, includingmotor neurons, circadian clock neurons, and the learning andmemory center neurons, and show in each case that minocyclinetreatment effectively restores synaptic connectivity architecturetowards that of wild type. We next genetically mimic the proposedMMP inhibition effect of minocycline treatment withoverexpression of the sole Drosophila TIMP in dfmr1 null mutants(Page-McCaw et al., 2007; Page-McCaw et al., 2003). Consistentwith the MMP hypothesis, we show that TIMP overexpression fullyrestores normal neuromuscular junction (NMJ) synapticarchitecture in the dfmr1 null mutant condition. Conversely, dfmr1removal similarly prevented TIMP overexpression phenotypes,including tracheal defects and lethality, suggesting that the TIMPand Drosophila FMRP (dFMRP) pathways are interdependent. InDrosophila, there are only two MMPs: secreted MMP1 andmembrane-anchored MMP2 (Page-McCaw et al., 2007; Page-McCaw et al., 2003). Drosophila mmp1 mutants exhibited both thetracheal defects and lethality that had been shown to be dFMRPdependent. We therefore generated dfmr1;mmp1 double nullmutants and similarly found the same mutual repression of bothclasses of null mutant phenotypes, which is consistent with theTIMP overexpression analyses. We did not detect changes inMMP1 expression or enzymatic activity in dfmr1 mutantscompared with controls, but did demonstrate striking geneticinteractions between dFMRP, TIMP and MMP1. These data suggestthat minocycline acts via MMP inhibition to alleviate FXS modeldefects, as previously reported (Bilousova et al., 2009), and providethe first direct proof for reciprocal, bidirectional interactionbetween the FMRP and TIMP molecular pathways.

RESULTSMinocycline partially restores dfmr1 null NMJ synapsemorphologyThe effects of dFMRP loss on NMJ architecture have been welldocumented (Coffee et al., 2010; Gatto and Broadie, 2008; Zhanget al., 2001). The dfmr1 null synapse displays an overgrowth defectwith increased branch number, more mature synaptic boutons andan accumulation of developmentally arrested satellite (or mini)boutons (Beumer et al., 2002; Beumer et al., 1999; Coffee et al.,2010; Gatto and Broadie, 2008). To explore the effect of minocyclinetreatment on these phenotypes, we assayed wandering third instarNMJs that were co-labeled with anti-horseradish-peroxidase (anti-HRP; to delineate the presynaptic terminal) and anti-discs-large(anti-DLG; to mark the postsynaptic compartment) (Fig. 1). Thenumber of NMJ branches, defined as a process with ≥2 boutons,was significantly higher in the dfmr1 null mutants compared withcontrol (Fig. 1A). The number of mature type 1b boutons, definedas DLG-positive varicosities ≥2 m in diameter, was likewisesignificantly increased in the dfmr1 null synapse (Fig. 1A,C). Thenumber of immature satellite boutons, defined as varicosities

synaptic bouton number in dfmr150M (see Zhang et al., 2001) nullmutants was significantly restored towards control levels (control:17.2±1.08, dfmr1: 31.4±1.71, P≤0.001; dfmr1 + minocycline:24.9±2.20, n12, P≤0.05; Fig. 1A,C). Moreover, the accumulationof developmentally arrested satellite boutons at the dfmr150M nullsynapse was fully restored to control levels (control: 1.9±0.25, dfmr1:4.1±0.39, P≤0.001; dfmr1 + minocycline: 1.83±0.30, n12), a highlysignificant effect (P≤0.001; Fig. 1B,D). To ensure that this effectwas not allele specific, we examined the effect of minocycline onan independent dfmr1 null allele (dfmr12) (see Dockendorff et al.,2002). Minocycline treatment again significantly restored defectsin both mature synaptic bouton number (dfmr12: 26.2±1.41; dfmr12+ minocycline: 18.8±2.62; n7; P≤0.05) and satellite boutonaccumulation (dfmr12: 3.43±0.43; dfmr12 + minocycline: 1.36±0.52;n7; P≤0.05) towards control levels. For both phenotypes, lowerdosages of minocycline were administered to determine dosagedependence. Clear evidence of a dose dependency was evident forboth mature boutons [in dfmr1 nulls: 30.5±2.64 (5 M), 29.7±1.04(10 M) and 24.9±2.20 (20 M); n12 each condition] andimmature boutons [in dfmr1 nulls: 3.86±0.40 (2 M), 2.85±0.33 (5M) and 1.83±0.30 (20 M); n12 each condition] (Fig. 1C,D). Nogreater restorative effects were observed in the dfmr1 null conditionat minocycline levels >20 M.

We conclude that minocycline can significantly reverse, in adosage-dependent manner, the increased synaptic bouton numberthat characterizes the dfmr1 null NMJ, particularly theaccumulation of developmentally arrested, immature satelliteboutons. By contrast, the increase in NMJ arbor branch numberin dfmr1 mutants was not significantly restored by minocyclinetreatment (control: 2.09±0.21, dfmr1: 3.08±0.22, P≤0.001; dfmr1 +minocycline (20 M): 2.92±0.22; n12 each condition). Withrespect to all these synaptic structural parameters, w1118 controlanimals treated with minocycline showed no significant differencecompared with untreated animals [control + minocycline: 17.1±1.76(type 1b synaptic bouton number), 1.75±0.39 (satellite boutonnumber) and 1.9±0.23 (branch number)]. Thus, minocyclineselectively rescued the synaptic bouton over-elaboration that

characterizes the FXS model condition without detectably alteringsynaptic architecture in the wild-type control.

Minocycline ameliorates dfmr1 null mutant circadian clock circuitdefectsFXS is associated with central synaptic dysfunction, and centralsynapses might respond differently to minocycline treatmentcompared with the peripheral NMJ. We therefore next wanted totest drug treatments in behaviorally relevant CNS circuits. Onecharacteristic behavioral manifestation in FXS patients is disruptedsleep patterns and associated hyperactivity (Elia et al., 2000;Penagarikano et al., 2007). Similarly, dfmr1 null animals exhibitcircadian arrhythmicity and hyperactivity during normal sleepperiods (Bushey et al., 2009; Dockendorff et al., 2002; Gatto andBroadie, 2009a; Gatto and Broadie, 2009b). In the underlyingcircadian clock circuit, a crucial subset of clock neurons expressingthe neuropeptide Pigment dispersing factor (PDF) show well-documented synaptic overgrowth defects in the dfmr1 nullcondition (Dockendorff et al., 2002; Gatto and Broadie, 2009a; Gattoand Broadie, 2009b; Inoue et al., 2002; Morales et al., 2002). Anti-PDF staining of these small ventrolateral (sLNv) clock neurons indfmr1 null brains revealed an increased number of synapticboutons, with overextension of their spatial distribution extendingfrom the dorsal horn bifurcation point (Fig. 2). We employed theseclock neuron synaptic architecture defects as the second test ofminocycline treatment effectiveness in the Drosophila FXS model.

We fed animals minocycline throughout larval development andadulthood. The same optimal minocycline concentration (20 M)was used as above for larval treatment. Adults eclosed on applejuice plates with yeast paste, both containing minocycline (1 mM).A higher concentration was used after eclosion because the drugis better tolerated in adults than in larvae. Staged animals wereaged to 3 days post-eclosion (d3) with brain dissections conductedat zeitgeber time 2-4 (2-4 hours after lights on). Consistent withearlier reports (Gatto and Broadie, 2009b), dfmr150M null brainsshowed increased numbers of sLNv clock neuron PDF-reactivesynaptic boutons in dorsal protocerebral projections (45±4.4

Disease Models & Mechanisms 675

Drosophila Fragile X minocycline trial RESEARCH ARTICLE

Fig. 1. Minocycline partially restores dfmr1 null NMJsynaptic morphology. (A)Representative images ofmuscle 4 NMJ synapse structure in control (w1118), dfmr1null (dfmr150M) and minocycline-treated (20M) dfmr1 nullwandering third instar larvae. Preparations were co-labeledwith postsynaptic anti-DLG (green) and presynaptic anti-HRP (red). Scale bar: 10m. (B)Higher magnificationimages showing NMJ synaptic boutons, with arrowheadsindicating satellite boutons. Scale bar: 5m.(C,D)Quantification of mature type 1b synaptic bouton (C)and satellite bouton (D) numbers. The minocyclineconcentrations are shown inM. Error bars indicate s.e.m.Sample size is n12 animals for each condition. Significanceis indicated as *P

control vs 67±6.3 dfmr1 null; n≥12; P

architecture. In young adult brains (0-4 hours post-eclosion),dfmr150M null Kenyon cell axonal length and branch number wereboth significantly increased (P

negative consequences of TIMP overexpression. We tested thisobservation by examining viability and cellular phenotypes in fourgenotypes: the driver-only control (UH1-Gal4/+), the ubiquitousTIMP overexpression condition (UH1-Gal4/UAS-TIMP), TIMPoverexpression in the dfmr150M heterozygous background (UH1-Gal4,dfmr1/UAS-TIMP) and TIMP overexpression in the dfmr150Mhomozygous background (UH1-Gal4,dfmr1/UAS-TIMP,dfmr1)(Fig. 5).

We first quantified the viability of these four genotypes (Fig. 5A).Control and experimental crosses were set to lay at 25°C at thesame time and were analyzed for viability at two time intervals: (1)from embryo (day 0) to pupation (day 5); and (2) from pupation

(day 5) to adulthood (day 10). TIMP overexpression caused reducedviability during the first phase (embryonic-larval development) andcomplete lethality during the second phase (pupal to adultdevelopment). Control animals showed 60% survival to pupation,whereas TIMP overexpression animals showed only 26% survival.Removal of dFMRP restored TIMP overexpression viability backto >50%. During pupation, nearly all control pupae eclosed to viableadults, whereas no adults overexpressing TIMP emerged (control:95.2±2.2% viable vs TIMP overexpression: 0.2±0.2%; n9 trials;P

removal of dFMRP can nearly completely restore the viability ofTIMP-overexpressing animals.

TIMP overexpression is also known to cause tracheal stretchingand deformation, usually resulting in dorsal tracheal breaks(Glasheen et al., 2009; Glasheen et al., 2010; Page-McCaw et al.,2003). Importantly, these tracheal defects phenocopy the mmp1null mutant, but not the mmp2 mutant, suggesting that MMP1 isspecifically involved in this mechanism. We therefore next testedthe effects of dFMRP loss on these tracheal TIMP overexpressionphenotypes (Fig. 5B). Animals overexpressing TIMP exhibited cleartracheal deformation, with these animals exhibiting 1.43±0.48breaks per dorsal trachea (n≥7) and with ~85% of trachea exhibitingat least one prominent tracheal break. By contrast, control animals(UH1-Gal4/+) never showed any detectable breaks in the dorsaltrachea (0 breaks; n≥7; Fig. 5B). Removal of one copy of dfmr150Mprovided no significant change in the TIMP overexpressionphenotype: these larvae had 1.14±0.34 breaks per dorsal trachea(n≥7; P>0.5 compared with UH1-Gal4/UAS-TIMP). By contrast,total dFMRP loss (homozygous dfmr150M) completely preventedTIMP overexpression tracheal defects (Fig. 5B). Animalsoverexpressing TIMP in the dfmr150M homozygous null backgroundnever showed any detectable breaks in the dorsal trachea (0 breaks;n≥7). Thus, removal of dFMRP can compensate for TIMPoverexpression defects at both the cellular level and at the level ofwhole animal viability. We conclude that TIMP and dFMRP mustreciprocally regulate each other in overlapping pathways.

dFMRP removal similarly prevents mmp1 null phenotypesIn Drosophila, there are only two MMPs: secreted MMP1 andmembrane-anchored MMP2 (Page-McCaw et al., 2007; Page-McCawet al., 2003). Drosophila mmp1 mutants exhibit both tracheal defectsand lethality that mimic the TIMP overexpression condition. Wetherefore next generated dfmr1;mmp1 double null mutants (Fig. 6).We confirmed the genotype with western blots for dFMRP andMMP1, showing both proteins to be completely absent in theanalyzed condition (Fig. 6A). As it did in TIMP overexpressionanimals, dFMRP removal rescued adult viability in mmp1 mutants,which alone showed 100% penetrant lethality (Fig. 6B). Null mmp1mutants had reduced viability during larval development comparedwith controls (w1118: 60%; mmp1: 17%). By sharp contrast, doublemutants were almost as viable as controls, with >50% during larvalsurvival to pupation. During pupation, the remnant mmp1 mutantsall died, whereas control animals almost all survived into adulthood(w1118: 94.3±3.32% viable vs mmp1: 0.33±0.29% viable; n8independent trials; P

MMP1 removal rescues dfmr1 null synaptic morphological defectsWestern blots show that, in the hippocampus of FMR1 KO mice(P7), active MMP9 expression levels were increased comparedwith controls, and MMP9 gelatinase activity was increased, asmeasured by gel zymography (Bilousova et al., 2009). Uponminocycline treatment, MMP9 expression and activity were bothreduced in the FMR1 KO mice compared with controls.Additionally, MMP9 treatment of wild-type hippocampal neuronsresulted in hippocampal dendritic spine profiles similar to FMR1KO spines. Thus, we wanted to first assay for biochemicaldifferences in MMP1 expression and activity in the dfmr1 nullnervous system. We first examined MMP1 expression in thenervous system by western blot using anti-MMP1 antibodies forthe catalytic domain (Glasheen et al., 2009). We found clearexpression of MMP1 in the nervous system, but no detectabledifference between control, dfmr150M heterozygous and dfmr150Mnull brains (supplementary material Fig. S1). Likewise, MMP1immunolabeling indicated the clear presence of secreted MMP1in the extracellular space surrounding NMJ synaptic boutons(supplementary material Fig. S2B), but no detectable change inexpression between controls and dfmr150M nulls uponquantification (supplementary material Fig. S2C). As a measureof gelatinase activity, we then employed a recently devised in situtechnique using a fluorescein-conjugated DQ-gelatin thatfluoresces when cleaved (Bajenaru et al., 2010; Vidal et al., 2010).We found clear indication of MMP-dependent enzymatic functionat the NMJ that was markedly decreased upon TIMPoverexpression (supplementary material Fig. S3A) and colocalizedwith MMP1 protein (compare to supplementary material Fig.S2B). We did not detect any obvious difference in gelatinaseactivity levels between control and dfmr150M null synapses(supplementary material Fig. S3B). The enzymatic readout offluorescence intensity quantification was not significantlydifferent between genotypes (supplementary material Fig. S3C).Thus, we conclude that MMP1 is appropriately positioned tointeract with dFMRP at the synapse, but we failed to detectchanges in MMP1 expression or activity in the dfmr1 nullcondition.

With clear evidence of MMP1 at the NMJ, we next employedour genetic double mutants to test the effect of MMP1 removal ondfmr1 null synapse architecture defects (Fig. 7). Representative NMJimages show w1118 controls, dfmr150M null mutants anddfmr1;mmp1 double mutants co-stained with anti-HRP (red,presynaptic) and anti-DLG (green, postsynaptic) (Fig. 7A). We againfound increases in synaptic bouton number (control: 15.69±0.87;dfmr1: 28.55±1.29; n≥21; P

previous work, among others, has characterized dfmr1 null synapticarchitecture defects in motor neurons at the NMJ (Gatto andBroadie, 2008; Repicky and Broadie, 2009; Zhang et al., 2001), inclock neurons of the circadian activity circuit (Gatto and Broadie,2009b), and in Kenyon cell neurons of the mushroom body learningand memory circuit (Pan et al., 2004; Tessier and Broadie, 2008).All of these defects were responsive to some degree to minocyclinetreatment. Interestingly, CNS circuit defects were more completelyrescued compared with the peripheral NMJ, although NMJ defectswere prevented in a dosage-dependent manner. Surprisingly, in allthree circuits, wild-type neuron morphology was not detectablyaffected by minocycline treatment, showing that loss of dFMRPsensitizes the cell structure to minocycline activity. By contrast,mouse studies have shown that minocycline alters wild-type neuronarchitecture (Bilousova et al., 2009), suggesting that a differencemust be present between Drosophila and mammalian systems. Weconclude that the Drosophila FXS model supports the mouse FXSmodel in suggesting that minocycline should be an effectivetreatment for FXS patients.

Prior to current clinical trials of mGluR inhibitors, pilot studieswere conducted in both Drosophila and mouse FXS models onlithium and MPEP, among other drugs (Berry-Kravis et al., 2008;Choi et al., 2010; Levenga et al., 2010; McBride et al., 2005;Meredith et al., 2010; Pan et al., 2008). Similarly, a very recent open-label add-on minocycline FXS treatment trial has been performed(Paribello et al., 2010). Encouragingly, the 19 FXS patients thatcompleted the study showed substantial improvement in four outof five measures on the Aberrant Behavior Checklist-Community(ABC-C) scales and a similar gain on the Clinical GlobalImprovement (CGI) scale. The trial reported a wide variety of FXSsymptoms being improved by minocycline treatment, includingirritability, stereotypy, hyperactivity and inappropriate speechsubscales (Paribello et al., 2010). Even though this trial was notdouble-blind or placebo-controlled, and was done on a small scale,these preliminary findings are highly promising. The widespreadbehavioral improvement in this trial is echoed in the present studyby the rescue upon minocycline treatment of three very differentneural circuits that function in locomotion, circadian activity andcognitive learning and memory. Another recent study showed thatthe most common side effect of minocycline treatment on FXSpatients was limited to gastrointestinal problems, including loss ofappetite (Utari et al., 2010). Taken together, these data supportmoving forward to full clinical trials of minocycline on FXSpatients. The fact that minocycline has long been an establisheddrug should greatly facilitate its development as a new FXStreatment (Kim and Suh, 2009).

Minocycline has been proposed to function as an MMP inhibitorin alleviating defects in the mouse FXS model (Bilousova et al.,2009). In FMR1 KO hippocampal lysates, both the active form ofMMP9 and its gelatinase activity were reportedly increased, andminocycline reduced MMP9 expression and, to a lesser extent,gelatinase activity. Additionally, MMP9 treatment of wild-typecultured hippocampal neurons caused dendritic spine defectsresembling FMR1 KO spines, mimicking a state of enhancedmGluR signaling and increased LTD, consistent with the mGluRtheory of FXS pathogenesis (Bilousova et al., 2009). It was thereforepostulated that enhanced mGluR signaling could be affectingincreased MMP expression and/or activity. To test this hypothesis

of MMP inhibition, we genetically overexpressed the soleDrosophila TIMP in a dfmr1 null animal, a direct genetic meansto inhibit MMP activity. The accessibility of the particularly well-characterized NMJ made it ideal for these mechanistic studies.Consistent with the hypothesis, we show complete prevention ofdfmr1 null defects in synaptic development. Moreover, to ourastonishment, we found reciprocal suppression of TIMPoverexpression phenotypes with regards to both cellular defectsand loss of viability. These data suggest that MMP inhibition mightbe responsible for the alleviatory effects of minocycline on FXSdefects. We further tested genetic double mutants that were nullfor both mmp1 and dfmr1. Consistent with the TIMPoverexpression studies, co-removal of dFMRP suppressed both thedevelopmental lethality and cellular defects of mmp1 mutants.Although no detectable change was observed in MMP1 expressionor enzymatic activity at the dfmr1 null NMJ, co-removal of mmp1significantly rescued dfmr1 synaptic architecture defects in amanner phenocopied by minocycline treatment. Thisindependently suggests that MMP1 inhibition might be theminocycline mechanism of action in the Drosophila FXS model.These data clearly support genetic interaction between TIMP anddFMRP pathways in the regulation of synaptic morphology.

The general reciprocal suppression of dfmr1 null with TIMP-overexpression and mmp1 null phenotypes suggests a strongrelationship between these pathways. The obvious question is todetermine how the two pathways interact, whether through director indirect means, or at the level of mRNA or protein. The moreplausible explanation is probably indirect interaction, perhaps inan mGluR-dependent mechanism at the synapse. In the mGluRtheory, receptor signaling drives elevated mRNA translationregulated by FMRP, which in turn controls levels of synapticglutamate receptors (GluRs) mediating LTD and LTP synapticplasticity (Bear et al., 2004). Mammalian MMP9 is secreted inresponse to GluR activation to mediate MMP-dependent LTD andLTP synaptic plasticity (Michaluk and Kaczmarek, 2007; Nagy etal., 2006; Tian et al., 2007). Like the Drosophila MMP1 expressionand activity reported here at the glutamatergic NMJ, MMP9 islocalized to mammalian glutamatergic synapses (Gawlak et al.,2009; Konopacki et al., 2007), although the mechanism of MMP9action in plasticity is still unclear. However, MMP synapticlocalization and involvement with mGluR-regulated synaptic eventsprovide a common foundation with FMRP function, from whichmechanistic studies can be launched. Potentially, FMRP couldcontrol many steps in the regulation of MMP release, localizationor activity regulation, and this control would not necessarily bereflected in gross changes in MMP expression or even basalenzymatic activity, consistent with the biochemical data in thisstudy. MMPs could reciprocally ameliorate FMRP activity bydirectly or indirectly modulating mGluR signaling. Although the‘mGluR theory’ provides a possible explanation for the interactionbetween MMP and FMRP at the synapse, the suppression of non-neuronal MMP phenotypes suggests that there also might be otherways in which these two pathways collide.

In closing, we stress that these studies provide strongcorrelation between the effects of minocycline treatment, TIMPoverexpression and MMP1 loss of function on the Drosophila FXSmodel, but a causal link in a common pathway has yet to beproven. Beyond MMPs, minocycline could also be affecting other

Disease Models & Mechanisms 681

Drosophila Fragile X minocycline trial RESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

targets to exert its effects. For example, minocycline is a knownp38 MAPK inhibitor (Kim and Suh, 2009). p38 MAPK has beenimplicated in mGluR-dependent LTD, which requires activationof p38 MAPK and protein tyrosine phosphatases in the CA1region of the adult rat hippocampus (Moult et al., 2008; Pi et al.,2004). The mGluR theory of FXS would allow for increased p38MAPK activation and increased tyrosine phosphatase activity tocause enhanced LTD, for example, which could be alleviated byminocycline inhibition. TIMPs also regulate this pathway. Bothmammalian TIMP-1 and TIMP-2 have been implicated in causingincreases in tyrosine kinase activity, the antagonist to the secondrequirement for p38-dependent enhanced LTD (Yamashita et al.,1996). Thus, the reciprocal suppression of phenotypes presentedhere could also support this type of model for minocycline activity.Additionally, in both amyotrophic lateral sclerosis andHuntington’s disease models, minocycline has been found to beeffective by inhibiting caspases and iNOS (Chen et al., 2000; Zhuet al., 2002). Thus, these two targets could also be viablemechanisms for minocycline action. Future studies will focus oninvestigating the links between minocycline activity and theTIMP pathway in the modulation of dFMRP requirements at thesynapse and in other tissues.

METHODSDrosophila geneticsAll stocks were maintained at 25°C on standard medium in a 12:12hour light-dark cycling humidified incubator. The geneticbackground strain was w1118. Two dfmr1 null allele stocks were used:dfmr150M/TM6GFP and dfmr12/TM6c (Zhang et al., 2001;Dockendorff et al., 2002). For MARCM analyses, the control wasobtained by crossing heatshock-FLP, mCD8-GFP; FRT82B, TubulinP-Gal80; Gal4-OK107 with y, w; FRT82B, and the null mutantcondition was obtained by crossing with FRT82B,dfmr150M/TM6GFP. Recombinant lines were generated with UAS-TIMP(Glasheen et al., 2009; Page-McCaw et al., 2003), with UH1-Gal4introduced into the dfmr150M null background using standard genetictechniques. G418 resistance and anti-dFMRP immunoblotting wereused to confirm the dfmr150M null background. For synaptic structureanalyses on the recombinant animals, the negative control wasobtained by crossing UH1-Gal4,dfmr150M/TM6GFP withdfmr150M/TM6GFP, and the experimental animals (kept strictly at25°C) were obtained through crossing UH1-Gal4,dfmr150M/TM6GFPwith UAS-TIMP,dfmr150M/TM6GFP. Double mutants weregenerated from mmp1Q112 null and dfmr150M null stocks usingstandard genetic techniques (Glasheen et al., 2009; Page-McCaw etal., 2003). Genotype was confirmed by western blots for dFMRP andMMP1 (Fig. 6A).

Minocycline administrationMinocycline (Sigma-Aldrich, St Louis, MO) was made as a 10 mMstock solution stored at 4°C. Genetic control and dfmr1 null strainswere maintained on apple juice agar plates with yeast paste in whichminocycline had been diluted at the specified concentrations inboth the agar and the yeast paste. Apple juice plates withminocycline were replaced every other day throughoutdevelopment with yeast paste containing minocycline made fresheach time. The same larval administration protocol was used forNMJ, clock neuron and MARCM assays. Adults were allowed to

eclose on fresh apple juice agar plates with yeast paste that bothcontained 1 mM minocycline.

Viability assaysDrosophila crosses were maintained on apple juice agar plates withyeast paste from egg laying (embryo) through adult eclosion, at25°C. Egg number was counted on the first laying day (day 0) andthe number of animals counted on each subsequent day. Todetermine viability until pupation {% viability [(pupae)/(embryos)]� 100}, the number of animals alive on day 5 was used forcalculations. For viability until adulthood {% viability [(adults)/(pupae)] � 100}, the number of adults on day 10 was used.

ImmunocytochemistryAntibody labeling was performed on third instar larvae and adultbrains as described previously (Coffee et al., 2010; Gatto andBroadie, 2008; Gatto and Broadie, 2009b; Tessier and Broadie, 2008).Briefly, samples were fixed in 4% paraformaldehyde in phosphatebuffered saline (PBS; pH 7.4) for 40 minutes. Preparations wererinsed with 3� PBS, and then blocked and permeabilized with 1%bovine serum albumin (BSA) and 0.5% normal goat serum (NGS)in 0.2% Triton X-100 in PBS (PBST) for 1 hour at room temperature.Primary and secondary antibodies were diluted in PBST with 0.2%BSA and 0.1% NGS, and incubated overnight at 4°C and for 2 hoursat room temperature, respectively. Primary antibodies usedincluded: anti-DLG [1:200; 4F3 monoclonal, Developmental StudiesHybridoma Bank (DSHB)], anti-HRP (1:250; rabbit, Sigma), anti-PDF (1:5; C7 monoclonal, DSHB), anti-GFP (1:500; clone 6662,FITC-conjugated goat, Abcam, Cambridge, MA), anti-FasII (1:10;1D4 monoclonal, DSHB) and mouse monoclonal cocktail anti-MMP1 to the catalytic domain (1:10; DSHB) (Glasheen et al., 2009;Page-McCaw et al., 2003). Secondary antibodies (1:200) used were:Alexa-Fluor-488-conjugated goat anti-mouse IgG, Alexa-Fluor-594-conjugated goat anti-rabbit IgG, Alexa-Fluor-488-conjugateddonkey anti-goat IgG and Alexa-Fluor-555-conjugated donkeyanti-mouse IgG (all from Invitrogen-Molecular Probes).Preparations were mounted in FluoroMount G (EMS, Hatfield, PA) and fluorescent images collected using a ZEISS LSM 510 META laser scanning confocal microscope. ImageJ(http://rsb.info.nih.gov/ij/) was used for fluorescence intensityquantifications.

NMJ structure analysesNMJs from wandering third instar larvae were quantified forsynaptic structure as previously described (Coffee et al., 2010; Gattoand Broadie, 2008). Briefly, the muscle 4 NMJ of abdominalsegment 3 was used for all quantification. Values were determinedfor both left and right hemisegments and averaged for each animal(n1). A synaptic branch was defined as an axonal projection withat least two synaptic boutons. Two populations of synaptic boutonswere defined for quantification: (1) type Ib (>2 m diameter) and(2) satellite or mini (≤2 m diameter and directly attached to a typeIb bouton). Each class of bouton is reported as number perterminal.

PDF clock neuron analysessLNv clock neuron structure was quantified as described previously(Coffee et al., 2010; Gatto and Broadie, 2009b). Briefly, a PDF-

dmm.biologists.org682

Drosophila Fragile X minocycline trialRESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

adapted Sholl array analysis was employed using a series of 10-m-spaced concentric rings centered on the dorsal horn bifurcation ofthe sLNv terminal arbor and extending for 100 m. Assays wereconducted at zeitgeber time 2-4. Both total number of PDF-positive synaptic boutons (≥1 m in diameter) and the number ofboutons within each concentric ring throughout the arbor werecounted. Values were determined for each arbor in both left andright hemispheres of the brain and averaged for each animal (n1).

MARCM analysesSingle-cell MARCM clones were generated and quantified aspreviously described (Lee and Luo, 1999; Pan et al., 2004; Tessierand Broadie, 2008). Briefly, to generate Kenyon cell clones in theMB -lobe, embryos were laid for a 4 hour period at 25°C and thenheat-shocked at 24 hours at 37°C for 1 hour. All quantification wasdone on single-cell -neuron MARCM clones. The axonal lengthand branch number parameters were determined with LSMsoftware on 3D confocal z-stacks for each individual neuron. For-neuron structural quantification, the primary axonal branch wasidentified first, with all other processes extending from the primaryaxon counted as branches. Axonal length was determined basedon the length of the primary axonal process.

Tracheal analysisTrachea were imaged in living animals using a Nikon 90i uprightcompound microscope connected to a Photometrics CoolSNAPHQ2 camera. Breaks in the dorsal trachea were counted peranimal for each assayed genotype.

Western immunoblottingWestern blots were performed as previously described (Coffee etal., 2010; Tessier and Broadie, 2008). Briefly, five brains weredissected per genotype and placed in 1� NuPage (Invitrogen,Carlsbad, CA) sample buffer with 2-mercaptoethanol. Samples werecentrifuged for 1 minute at 12,000 g, boiled for 10 minutes andcentrifuged again for 1 minute. Samples were then loaded onto a4-12% Bis-Tris gel, electrophoresed at 200 V for 50 minutes, andtransferred at 100 V for 1 hour to nitrocellulose. Membranes wererinsed with NanoPure water, blocked in Odyssey blocking buffer(Li-Cor, Lincoln, NE) for 1 hour, and probed overnight withprimary antibody at 4°C. Blots were then washed three times in0.1% PBST for 5 minutes, incubated with secondary antibody at25°C for 1 hour, and washed in a similar manner for 5 minutesthree times each. Primary antibodies used were: mouse monoclonalcocktail anti-MMP1 to the catalytic domain (DSHB) (Glasheen etal., 2009; Page-McCaw et al., 2003), mouse monoclonal anti-dFMRP (1:2500, DSHB) and mouse monoclonal anti-actin (JLA20,1:50; DSHB). Secondary antibodies were: Alexa-Fluor-680-conjugated goat anti-mouse (1:10,000) and IRDye-800 goat anti-mouse (1:1000). Odyssey software was used for image capture andanalysis.

In vivo zymographyThe in situ zymography technique was adapted from recent studiesin mice and Drosophila (Bajenaru et al., 2010; Vidal et al., 2010).Briefly, wandering third instars were dissected and incubatedimmediately in 500 g/ml of DQ-gelatin (Fluorescein conjugated;Molecular Probes) for 40 minutes in 1� reaction buffer (Molecular

Probes) at 25°C. Preparations were rinsed in PBS three times andthen fixed for 30 minutes in 4% paraformaldehyde/4% sucrose inPBS at 25°C. After three PBS washes, preparations were incubatedwith Cy3-conjugated anti-HRP (1:200, goat; JacksonImmunoResearch Laboratories, West Grove, PA) for 2 hours at25°C. Samples were washed three times in PBS and mounted inFluoroMount G (EMS, Hatfield, PA). All fluorescent images were

Disease Models & Mechanisms 683

Drosophila Fragile X minocycline trial RESEARCH ARTICLE

TRANSLATIONAL IMPACT

Clinical issueFragile X syndrome (FXS), which occurs in ~1 in 4000 males and ~1 in 6000females, causes a variety of behavioral symptoms, including hyperactivity,disrupted circadian patterns, and cognitive learning and memory deficits. FXSresults from a loss of the fragile X mental retardation 1 (FMR1) protein product(FMRP), which is an mRNA-binding translational regulator. Loss of FMRP isproposed to cause enhanced metabotropic glutamate receptor (mGluR)signaling, which leads to increased long-term potentiation (LTP) anddepression (LTD) (affecting synaptic plasticity). Tests in mouse and Drosophilamodels of FXS suggest that mGluR inhibitors are an effective means by whichto prevent or rescind FXS phenotypes, but developing other methods oftreatment is also desired. A recent mouse study indicated that the tetracyclinederivative minocycline might be an effective therapy for FXS through its abilityto inhibit upregulated matrix metalloproteinase-9 (MMP9), a secretedgelatinase that cleaves extracellular proteins to remodel the extracellularmatrix.

ResultsIn this study, the authors use the well-characterized Drosophila FXS diseasemodel (dfmr1 null) to test the effects of minocycline in three different classesof neural circuits: the neuromusculature, the circadian clock circuit, and thebrain learning and memory center. In all three locations, they find that mutantsynaptic connectivity defects are rescued by minocycline treatment. To testthe MMP hypothesis, they overexpress the sole Drosophila tissue inhibitor ofmatrix metalloproteinases (TIMP) in dfmr1 null mutant flies and assay forrescue of synaptic defects. Similarly to minocycline, TIMP overexpression alsorestores normal synaptic connectivity in the Drosophila FXS model. Moreover,removal of Drosophila FMRP (dFMRP) reciprocally prevents defects induced byTIMP overexpression, including cellular phenotypes and pre-adult lethality,showing that the TIMP and dFMRP pathways bidirectionally overlap. Similarly,removal of dFMRP suppresses both lethality and tracheal breaking defectsobserved in mmp1 null flies. Despite the fact that MMP1 expression andgelatinase activity in the nervous system are found to be the same in controland dfmr1 null flies, removal of MMP1 from dfmr1 null flies partially rescuesFXS synaptic structure defects. These data are consistent with the MMPhypothesis and explain how minocycline suppresses defects in animal modelsof FXS.

Implications and future directionsThe results of the recent mouse study and this Drosophila study concur thatminocycline is a highly effective treatment in FXS disease models. Moreover, avery recent open-label, add-on clinical trial on FXS patients also stronglysupports the therapeutic potential of minocycline. Together, these studiesmandate that minocycline should be tested in a double-blind, placebo-controlled clinical trial in patients with FXS. Additionally, the reciprocalregulation of the FMRP and TIMP pathways found here suggest that morework is necessary to further elucidate the interplay between these pathwaysand the common components of each that might be targeted in noveltreatment strategies for FXS. Conversely, this study also suggests thatinvestigating FMRP and its targets in cancer biology and therapeutics iswarranted, given the central role of MMPs in cancer progression.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

collected using a ZEISS LSM 510 META laser scanning confocalmicroscope. ImageJ was used for fluorescence intensityquantification.

StatisticsStatistical analyses were performed using GraphPad InStat 3(GraphPad Software, San Diego, CA). The more conservativeunpaired, nonparametric, two-tailed Mann-Whitney tests wereapplied to determine significance. For drug treatment assays,means of control versus dfmr1 nulls were compared, and meansof dfmr1 nulls versus each drug-treated condition were compared.For genetic analyses, the control animal was compared with themutant condition, which was then compared with rescue animals.All other comparisons were control to dfmr1 nulls. Significance isrepresented in the figures as P

Konopacki, F. A., Rylski, M., Wilczek, E., Amborska, R., Detka, D., Kaczmarek, L.and Wilczynski, G. M. (2007). Synaptic localization of seizure-induced matrixmetalloproteinase-9 mRNA. Neuroscience 150, 31-39.

Koukoui, S. D. and Chaudhuri, A. (2007). Neuroanatomical, molecular genetic, andbehavioral correlates of fragile X syndrome. Brain Res. Rev. 53, 27-38.

Laggerbauer, B., Ostareck, D., Keidel, E. M., Ostareck-Lederer, A. and Fischer, U.(2001). Evidence that fragile X mental retardation protein is a negative regulator oftranslation. Hum. Mol. Genet. 10, 329-338.

Lee, T. and Luo, L. (1999). Mosaic analysis with a repressible cell marker for studies ofgene function in neuronal morphogenesis. Neuron 22, 451-461.

Levenga, J., de Vrij, F. M., Oostra, B. A. and Willemsen, R. (2010). Potentialtherapeutic interventions for fragile X syndrome. Trends Mol. Med. 16, 516-527.

Li, Z., Zhang, Y., Ku, L., Wilkinson, K. D., Warren, S. T. and Feng, Y. (2001). The fragileX mental retardation protein inhibits translation via interacting with mRNA. NucleicAcids Res. 29, 2276-2283.

Lu, R., Wang, H., Liang, Z., Ku, L., O’Donnell, W. T., Li, W., Warren, S. T. and Feng, Y.(2004). The fragile X protein controls microtubule-associated protein 1B translationand microtubule stability in brain neuron development. Proc. Natl. Acad. Sci. USA101, 15201-15206.

Margulies, C., Tully, T. and Dubnau, J. (2005). Deconstructing memory in Drosophila.Curr. Biol. 15, R700-R713.

McBride, S. M., Choi, C. H., Wang, Y., Liebelt, D., Braunstein, E., Ferreiro, D., Sehgal,A., Siwicki, K. K., Dockendorff, T. C., Nguyen, H. T. et al. (2005). Pharmacologicalrescue of synaptic plasticity, courtship behavior, and mushroom body defects in aDrosophila model of fragile X syndrome. Neuron 45, 753-764.

Meredith, R. M., de Jong, R. and Mansvelder, H. D. (2010). Functional rescue ofexcitatory synaptic transmission in the developing hippocampus in Fmr1-KO mouse.Neurobiol. Dis. 41, 104-110.

Michaluk, P. and Kaczmarek, L. (2007). Matrix metalloproteinase-9 in glutamate-dependent adult brain function and dysfunction. Cell Death Differ. 14, 1255-1258.

Morales, J., Hiesinger, P. R., Schroeder, A. J., Kume, K., Verstreken, P., Jackson, F. R.,Nelson, D. L. and Hassan, B. A. (2002). Drosophila fragile X protein, DFXR, regulatesneuronal morphology and function in the brain. Neuron 34, 961-972.

Moult, P. R., Correa, S. A., Collingridge, G. L., Fitzjohn, S. M. and Bashir, Z. I. (2008).Co-activation of p38 mitogen-activated protein kinase and protein tyrosinephosphatase underlies metabotropic glutamate receptor-dependent long-termdepression. J. Physiol. 586, 2499-2510.

Muddashetty, R. S., Kelic, S., Gross, C., Xu, M. and Bassell, G. J. (2007). Dysregulatedmetabotropic glutamate receptor-dependent translation of AMPA receptor andpostsynaptic density-95 mRNAs at synapses in a mouse model of fragile Xsyndrome. J. Neurosci. 27, 5338-5348.

Nagy, V., Bozdagi, O., Matynia, A., Balcerzyk, M., Okulski, P., Dzwonek, J., Costa, R.M., Silva, A. J., Kaczmarek, L. and Huntley, G. W. (2006). Matrix metalloproteinase-9 is required for hippocampal late-phase long-term potentiation and memory. J.Neurosci. 26, 1923-1934.

Page-McCaw, A. (2008). Remodeling the model organism: matrix metalloproteinasefunctions in invertebrates. Semin. Cell Dev. Biol. 19, 14-23.

Page-McCaw, A., Serano, J., Sante, J. M. and Rubin, G. M. (2003). Drosophila matrixmetalloproteinases are required for tissue remodeling, but not embryonicdevelopment. Dev. Cell 4, 95-106.

Page-McCaw, A., Ewald, A. J. and Werb, Z. (2007). Matrix metalloproteinases and theregulation of tissue remodelling. Nat. Rev. Mol. Cell Biol. 8, 221-233.

Pan, L. and Broadie, K. S. (2007). Drosophila fragile X mental retardation protein andmetabotropic glutamate receptor A convergently regulate the synaptic ratio ofionotropic glutamate receptor subclasses. J. Neurosci. 27, 12378-12389.

Pan, L., Zhang, Y. Q., Woodruff, E. and Broadie, K. (2004). The Drosophila fragile Xgene negatively regulates neuronal elaboration and synaptic differentiation. Curr.Biol. 14, 1863-1870.

Pan, L., Woodruff, E., 3rd, Liang, P. and Broadie, K. (2008). Mechanistic relationshipsbetween Drosophila fragile X mental retardation protein and metabotropicglutamate receptor A signaling. Mol. Cell. Neurosci. 37, 747-760.

Paribello, C., Tao, L., Folino, A., Berry-Kravis, E., Tranfaglia, M., Ethell, I. M. andEthell, D. W. (2010). Open-label add-on treatment trial of minocycline in fragile Xsyndrome. BMC Neurol. 10, 91.

Penagarikano, O., Mulle, J. G. and Warren, S. T. (2007). The pathophysiology offragile x syndrome. Annu. Rev. Genomics Hum. Genet. 8, 109-129.

Pi, R., Li, W., Lee, N. T., Chan, H. H., Pu, Y., Chan, L. N., Sucher, N. J., Chang, D. C., Li,M. and Han, Y. (2004). Minocycline prevents glutamate-induced apoptosis ofcerebellar granule neurons by differential regulation of p38 and Akt pathways. J.Neurochem. 91, 1219-1230.

Pieretti, M., Zhang, F. P., Fu, Y. H., Warren, S. T., Oostra, B. A., Caskey, C. T. andNelson, D. L. (1991). Absence of expression of the FMR-1 gene in fragile Xsyndrome. Cell 66, 817-822.

Popovic, N., Schubart, A., Goetz, B. D., Zhang, S. C., Linington, C. and Duncan, I. D.(2002). Inhibition of autoimmune encephalomyelitis by a tetracycline. Ann. Neurol.51, 215-223.

Repicky, S. and Broadie, K. (2009). Metabotropic glutamate receptor-mediated use-dependent down-regulation of synaptic excitability involves the fragile X mentalretardation protein. J. Neurophysiol. 101, 672-687.

Rivera, S., Khrestchatisky, M., Kaczmarek, L., Rosenberg, G. A. and Jaworski, D. M.(2010). Metzincin proteases and their inhibitors: foes or friends in nervous systemphysiology? J. Neurosci. 30, 15337-15357.

Rudelli, R. D., Brown, W. T., Wisniewski, K., Jenkins, E. C., Laure-Kamionowska, M.,Connell, F. and Wisniewski, H. M. (1985). Adult fragile X syndrome. Clinico-neuropathologic findings. Acta Neuropathol. 67, 289-295.

Stetler-Stevenson, W. G. (2008). Tissue inhibitors of metalloproteinases in cellsignaling: metalloproteinase-independent biological activities. Sci. Signal. 1, re6.

Tessier, C. R. and Broadie, K. (2008). Drosophila fragile X mental retardation proteindevelopmentally regulates activity-dependent axon pruning. Development 135,1547-1557.

Tessier, C. R. and Broadie, K. (2010). The fragile X mental retardation proteindevelopmentally regulates the strength and fidelity of calcium signaling inDrosophila mushroom body neurons. Neurobiol. Dis. 41, 147-159.

Tian, L., Stefanidakis, M., Ning, L., Van Lint, P., Nyman-Huttunen, H., Libert, C.,Itohara, S., Mishina, M., Rauvala, H. and Gahmberg, C. G. (2007). Activation ofNMDA receptors promotes dendritic spine development through MMP-mediatedICAM-5 cleavage. J. Cell Biol. 178, 687-700.

Utari, A., Chonchaiya, W., Rivera, S. M., Schneider, A., Hagerman, R. J., Faradz, S.M., Ethell, I. M. and Nguyen, D. V. (2010). Side effects of minocycline treatment inpatients with fragile X syndrome and exploration of outcome measures. Am. J.Intellect. Dev. Disabil. 115, 433-443.

Vidal, M., Salavaggione, L., Ylagan, L., Wilkins, M., Watson, M., Weilbaecher, K.and Cagan, R. (2010). A role for the epithelial microenvironment at tumorboundaries: evidence from Drosophila and human squamous cell carcinomas. Am. J.Pathol. 176, 3007-3014.

Wang, L. W., Berry-Kravis, E. and Hagerman, R. J. (2010). Fragile X: leading the wayfor targeted treatments in autism. Neurotherapeutics 7, 264-274.

Wang, X., Zhu, S., Drozda, M., Zhang, W., Stavrovskaya, I. G., Cattaneo, E.,Ferrante, R. J., Kristal, B. S. and Friedlander, R. M. (2003). Minocycline inhibitscaspase-independent and -dependent mitochondrial cell death pathways in modelsof Huntington’s disease. Proc. Natl. Acad. Sci. USA 100, 10483-10487.

Waung, M. W. and Huber, K. M. (2009). Protein translation in synaptic plasticity:mGluR-LTD, Fragile X. Curr. Opin. Neurobiol. 19, 319-326.

Wu, D. C., Jackson-Lewis, V., Vila, M., Tieu, K., Teismann, P., Vadseth, C., Choi, D. K.,Ischiropoulos, H. and Przedborski, S. (2002). Blockade of microglial activation isneuroprotective in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mouse modelof Parkinson disease. J. Neurosci. 22, 1763-1771.

Yamashita, K., Suzuki, M., Iwata, H., Koike, T., Hamaguchi, M., Shinagawa, A.,Noguchi, T. and Hayakawa, T. (1996). Tyrosine phosphorylation is crucial for growthsignaling by tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2). FEBS Lett.396, 103-107.

Zars, T., Fischer, M., Schulz, R. and Heisenberg, M. (2000). Localization of a short-term memory in Drosophila. Science 288, 672-675.

Zhang, Y. Q. and Broadie, K. (2005). Fathoming fragile X in fruit flies. Trends Genet. 21,37-45.

Zhang, Y. Q., Bailey, A. M., Matthies, H. J., Renden, R. B., Smith, M. A., Speese, S. D.,Rubin, G. M. and Broadie, K. (2001). Drosophila fragile X-related gene regulates theMAP1B homolog Futsch to control synaptic structure and function. Cell 107, 591-603.

Zhu, S., Stavrovskaya, I. G., Drozda, M., Kim, B. Y., Ona, V., Li, M., Sarang, S., Liu, A.S., Hartley, D. M., Wu, D. C. et al. (2002). Minocycline inhibits cytochrome c releaseand delays progression of amyotrophic lateral sclerosis in mice. Nature 417, 74-78.

Disease Models & Mechanisms 685

Drosophila Fragile X minocycline trial RESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

SUMMARYIntroductionRESULTSMinocycline partially restores dfmr1 null NMJ synapse morphologyMinocycline ameliorates dfmr1 null mutant circadian clock circuit defectsMinocycline restores dfmr1 null mushroom body neuron defectsTIMP overexpression fully rescues dfmr1 null synaptic morphological defectsdFMRP removal reciprocally suppresses TIMP overexpression phenotypesdFMRP removal similarly prevents mmp1 null phenotypesMMP1 removal rescues dfmr1 null synaptic morphological defects

Fig. 1.Fig. 2.Fig. 3.Fig. 4.Fig. 5.Fig. 6.DISCUSSIONFig. 7.METHODSDrosophila geneticsMinocycline administrationViability assaysImmunocytochemistryNMJ structure analysesPDF clock neuron analysesMARCM analysesTracheal analysisWestern immunoblottingIn vivo zymographyStatistics

TRANSLATIONAL IMPACTSupplementary materialREFERENCES