Brain-Derived Neurotrophic Factor Val66Met Allele Impairs Basal and Ketamine-Stimulated Synaptogenesis in Prefrontal Cortex

Download Brain-Derived Neurotrophic Factor Val66Met Allele Impairs Basal and Ketamine-Stimulated Synaptogenesis in Prefrontal Cortex

Post on 23-Nov-2016




0 download

Embed Size (px)


  • B FIm -SS l CRo mbGe

    Ba neutra iventra ice wket ogen

    Me ells inan in re

    Re tropexc addiim t micon c trantha was

    Co F Msyn ng thblo e.


    Asioto a valine-to-methionine substitution in the proBDNF protein atcoderacithadedetradeVahuthitio


    curmiofactresdebemilation of BDNF expression (17). It remains to be determined




    00doon 66 (Val66Met), is carried by approximately 30% of the gen-l population and has been associated with mild cognitive defi-s (57). In addition, individuals who carry the Met polymorphismvebeen reported tohave an increased risk of stress-relatedmajorpression (810). The Val66Met polymorphism impairs activity-pendent secretion of BDNF at synaptic sites (5,11) and reducesnslin-mediated trafficking of BDNF messenger RNA (mRNA) tondrites (12). Recently, Chen et al. (13) generated a BDNFl66Met knock-in mouse exhibiting phenotypic hallmarks of thisman polymorphism, providing an animal model for analysis ofs human loss-of-function BDNF polymorphism on synaptic func-n.In the present study, we focused on the mPFC because of itsotal role in neurocircuits underlying major depression (1416).viously we found that rats exposed to repeated mild restraintess or the stress hormone corticosterone exhibit atrophyof distalical dendrites and matching deficits in excitatory postsynaptic

    whether mice with the reduced-function BDNF Val66Met knock-inwould exhibit similar synaptic deficits.

    Recently, it was reported that a single subanesthetic dose of theshort-acting N-methyl-D-aspartate (NMDA) antagonist ketaminerapidly increases the density of functional synaptic spines in mPFC.This synaptogenic effect is preceded by a transient increase inmammalian Target of Rapamycin (mTOR) inprefrontal synaptoneu-rosomes followed within approximately 2 hours by a prolongedincrease in synaptic proteins and a concurrent antidepressant ac-tion lasting 1 week or more (22). These effects are blocked byrapamycin (intracerebroventricular), a drug that disrupts activationofmTOR, a serine/threonine kinase instrumental in the activationofdendritic translation of synaptic proteins (23). Studies in isolateddendrites demonstrate that puff application of BDNF increases lo-cal translation of proteins in an anatomically restricted, mTOR-de-pendent manner (24). The BDNF Val66Met knock-in mouse offersan opportunity to test whether this polymorphism, by impairingBDNF function, interferes with the synaptogenic and antidepres-sant effects of ketamine.

    Methods andMaterials

    AnimalsThe generation of the BDNF Val66Met knock-in mice has been

    previously described (13). Heterozygous Met/Val mice were bred,and the 3 resulting littermate genotypes, Val/Val wildtype (WT),

    m the Departments of Psychiatry (R-JL, X-YL, FB, RSD, GKA) and Pharma-cology (RSD, GKA), Yale School of Medicine, New Haven, Connecticut;and the Department of Psychiatry (FSL), Weill Cornell Medical College ofCornell University, New York, New York.dress correspondence to George Aghajanian, M.D., Ph.D., Yale School ofMedicine, Departments of Psychiatry and Pharmacology, 34 Park Street,New Haven, CT 06508; E-mail: Aug 11, 2011; revised Sep 27, 2011; accepted Sep 28, 2011.

    BIOL PSYCHIATRY 2012;71:996100506-3223/$36.00i:10.1016/j.biopsych.2011.09.030 2012 Society of Biological Psychiatryrain-Derived Neurotrophicpairs Basal and Ketamine

    ynaptogenesis in Prefrontang-Jian Liu, Francis S. Lee, Xiao-Yuan Li, Francis Baorge K. Aghajanian

    ckground: Knock-in mice with the common human brain-derivedffickingof BDNFmessengerRNA todendrites. Itwashypothesized, gnslation of BDNFmessenger RNA, that loss-of-function Met allele mamine, an N-methyl-D-aspartate antagonist that stimulates synapt

    thods: Whole-cell recordings from layer Vmedial PFC pyramidal calysis of wildtype, Val/Met, and Met/Met mice both at baseline and

    sults: Val/Met andMet/Metmice were found to have constitutive aitatory postsynaptic currents in layer V pyramidal cells of PFC. Inpaired synaptic formation/maturation (synaptogenesis). In Met/Mesistentwith the idea that synaptogenesis is dependent on dendritit the antidepressant response to ketamine in the forced swim test

    nclusions: The results demonstrate that expression of the BDNaptogenic and antidepressant actions of ketamine in PFC, suggesticked in depressed patients who carry the loss of function Met allel

    y Words: Antidepressant, dendritic spines, EPSC, hypocretin,jor depression, N-methyl-D-aspartate, serotonin

    decrease in brain-derived neurotrophic factor (BDNF) ex-pression in medial prefrontal cortex (mPFC) and other re-gionshasgiven rise to theBDNFhypothesis ofmajordepres-

    n (14). A human polymorphism in the BDNF gene, which leadsactor Val66Met Alleletimulatedortex

    ico, Ronald S. Duman, and

    rotrophic factor (BDNF) Val66Met polymorphism have impairedevidence that local synapse formation is dependentondendriticould show synaptic deficits both at baseline and in response toesis in prefrontal cortex (PFC).

    brain slices were combinedwith two-photon laser scanning forsponse to a low dose of ketamine.

    hy of distal apical dendrites and decrements in apically targetedtion, spine density and diameter were decreased, indicative ofce the synaptogenic effect of ketamine was markedly impaired,slation/release of BDNF. In parallel behavioral studies, we foundblocked in Met/Met mice.

    et allele in mice results in basal synaptic deficits and blocksat the therapeutic response to this drugmight be attenuated or

    rents (EPSCs) generated in the apical dendrites of layer V pyra-dal neurons of the mPFC (17). The synaptic responses consistedEPSCs elicited by serotonin (5-HT) and hypocrtin/orexin, whichvia apically targeted corticocortical and thalamocortical inputs,pectively (18,19). Given the role of BDNF in modulating activity-pendent synaptic plasticity amongmatureneurons (20,21), it hasen suggested that reductions in EPSCs induced by chronic stressght be mediated by a stress/corticosterone-induced downregu-

  • Val/Met, and Met/Met mice were used. The age of mice was 68momeholigtiohoduguUs


    anwitteeinin w(suPFACBabetioransliceq12mmmmenaprec




    resting potential (approximately 75 5mV) tominimize holdingcurtostovidCa




    tramocmtemmiJapthemiThmaindspeeaca hiogint


    R.-J. Liu et al. BIOL PSYCHIATRY 2012;71:9961005 997nths for the electrophysiology and spine morphology experi-nts and 310 months for the behavioral studies. Mice wereused and maintained in standard conditions with a 12-hourht/dark cycle and ad libitum access to food and water. Injec-ns of ketamine hydrochloride (10 mg/kg, IP) were made 24urs before preparation of brain slices. Animal use and proce-res were in accordance with the National Institutes of Healthidelines and approved by the Yale University Animal Care ande Committees.

    in Slice PreparationBrain slices were prepared as described (17). Briefly, mice wereesthetized with chloral hydrate (400 mg/kg, IP), in adherenceh protocols approved by the Yale Animal Care and Use Commit-. After decapitation, thebrainswere removed rapidly andplacedice-cold (approximately 4C) artificial cerebrospinal fluid (ACSF)hich sucrose (252mmol/L) was substituted for sodium chloride

    crose-ACSF) to prevent cell swelling. A block of tissue containingCwas dissected, and coronal slices (400m)were cut in sucrose-SF with an oscillating-blade tissue slicer (Leica VT1000S; Leica,nnockburn, Illinois) and placed in a submerged recording cham-r; bath temperaturewas raised slowly to 32C. Known concentra-ns of drugs dissolved in ACSF, applied through a stopcock ar-gement at a fast flow rate (approximately 4ml/min), reached theewithin 710 s. The standard ACSF (pH approximately 7.35) wasuilibrated with 95% oxygen/5% carbon dioxide and contained8 mmol/L sodium chloride, 3 mmol/L potassium chloride, 2ol/L calcium chloride, 2 mmol/L magnesium sulphate, 24ol/L sodium bicarbonate, 1.25 mmol/L sodium dihydrog-

    orthophosphate, and 10mmol/L D-glucose. A recovery period ofproximately 12 hours was allowed before commencement ofording.

    ctrophysiologyPyramidal neurons in layer V were visualized by an Olympus50WI microscope (40 or60 infrared [IR] lens) with IR differ-tial interference contrast (IR/differential interference con-st) video-microscopy (Olympus, Hamburg, Germany), as de-ibed (19). Low-resistance patch pipettes (35 M) werelled from patch-clamp glass tubing (Warner Instruments,mden, Connecticut) by a Flaming-Brown Horizontal Pullerodel P-97; Sutter Instruments, Novato, California). Pipettesre filled with the following solution: 115 mmol/L potassiumconate, 5 mmol/L potassium chloride, 2 mmol/L magnesiumloride, 2mmol/Lmagnesium-adenosine triphosphate, 2 mmol/Lodium (Na2) adenosine triphosphate, 10 mmol/L Na2-phos-ocreatine, .4 mmol/L Na2 guanosine triphosphate, and 10ol/L Hepes, pH 7.33. Neurobiotin (.3%) was added to theette solution to mark cells for later imaging. Pipettes weret tip-filled with regular patch solution before backfilling withNeurobiotin solution to avoid ejecting excess dye into the

    tracellular space of the slice.Whole-cell recordings were made with an Axoclamp-2B am-fier (Axon Instruments, Union City, California). The outputnal was low-pass-filtered at 3 KHz, amplified 100 throughberamp, digitized at 15 kHz, and acquired with pClamp 9.2/idata 1320 software (Axon Instruments). Series resistance,nitored throughout the experiment, was usually between 4d 8 M. To minimize series resistance errors, cells were dis-ded if series resistance rose above 10. Postsynaptic currentsre in continuous single-electrode voltage-clampmode (3-kHz-pass filter cutoff frequency); cells were clamped near theirrents. After completion of recording, slices were transferred4% paraformaldehyde in .1 mol/L phosphate buffer andred overnight at 4C. Slices were then processed with strepta-in conjugated to Alexa 594 (1/1000; Invitrogen, Carlsbad,lifornia) for Neurobiotin visualization.

    aging and Data AnalysisLabeled neurons within layer V of anterior cingulate (Cg1)d prelimbic mPFC (Cg3) were imaged with a two-photon lasernning system consisting of a Ti:sapphire laser (Mai Tai; Spec-Physics, Santa Clara, California) tuned to wavelength 810 nmd a direct detection Bio-Rad Radiance 2100 MP laser scanneriss Microimaging, Thornwood, New York) mounted on anmpus BX50WI microscope with 40 (.8 numerical aperture)60 (.9 numerical aperture) water-immersion objectivesympus). Total apical and basal dendritic branch length, 3-di-nsional Z-stacks were reconstructed from approximately 400 sequential scans at low zoom (312 312 m; 40 lens) atm steps so as to include the entire apical tuft and basalnches for a given cell within the slice. Total branch lengthwastermined within the 3-dimensional matrix of each Z-stackh Neurolucida 9 (MicroBrightField, Williston, Vermont). Spinensity, spine head diameter, and spine length analysis werene with Neuolucida Explorer (version 9) on the raw imagecks (29 optical sections, 1 m apart). Spine density waspled in two zones: 1) tips of apical tuft branches as they

    proach the pial membrane, and 2) proximal tuft dendrites justtal to the bifurcation. Results were expressed in terms of spinensity/10 m.

    rced Swim TestThe mouse forced swim test (FST) examines the dynamics ofnsition from an active (struggling) to passive (immobility)de of coping in an inescapable water-filled glass beaker (11diameter, 15.5 cm high; 10 cm water depth; 25C waterperature). In each run, four mice were video-recorded for 6

    n with a digital video camcorder (DCR-SX20, Sony, Tokyo,an). The locations of the animals were counter-balanced onbasis of their experimental groups. Over the first 2 min, test

    ce normally undergo habituation of struggling behavior.ereafter, the antidepressant or pro-depressive effects of phar-cological and genetic manipulations could be distinguished,icated by an enhancement of struggling and immobility, re-ctively, with manual (25) or automated scoring (26). Afterh run, mice were rescued, dried with a towel, and placed neareat source. All runs were conducted under minimal anx-enic conditions (27). Video clips were stored as audio videoerleaved (AVI) files for offline analyses.For quantification of animal movement from the FST videos, silhouette alteration method was employed with an auto-ted behavioral tracking system (Videotrack; View Point Lifeence, Montreal, Canada). Each video image is rapidly scannedint-by-point and line-after-line under an 800 600 pixel set-g (for 40 msec, [i.e., 25 images/sec]). Each point is assigned ainosity value of 0255 on a gray scale. Values were thenverted to a binary image: each pixel is white (0) or black (1).

    ce were distinguished from the background by adjusting thesitivity threshold of this conversion, associating all non-zeroels in each frame with the silhouette of the mouse (includinghead, tail, and paws). During movement analyses, degree ofouette alteration (i.e., the number of pixels coded 0 or 1)nged between successive frames, indicates the level of

  • gling or immobility displayed. With an integrated threshold al-go20imsuchalim


    Fig otropatr showVal ch lenden p .0Me dMeof b ritic sin V

    998 BIOL PSYCHIATRY 2012;71:9961005 R.-J. Liu et al.

    wwrithm, the threshold for immobility is fine-tuned (at a value of) before the actual analyses so that the mouse is consideredmobile (minimal movements) when it is visually verified ash. This and other automated FST behavioral quantitationve been tested to serve as reliable methods that address theitations of manual (visual) scoring by experimenters (26,28).

    ure 1.Medial prefrontal cortex layer V pyramidal cells in brain-derived neurophy. (A) Apical dendrites: (a) representative images of Z-stack projections/Val, Val/Met, andMet/Met mice; (b) bar graph comparing decrease in brandrites; oblique brancheswere decreased inMet/Metmice only. *p .05; **t allele mice (Z-stack projections of basal dendrites from Val/Val, Val/Met, anasal dendrites in Val/Met andMet/Metmice; (c) Sholl analyses of basal dendal/Met and Met/Met mice between 20 and 100 m from the hydrochloride was from Hospira (Lake Forest, Illi-is). For the slice experiments, 5-hydroxytryptamine, creatininefate (5-HT) was from Sigma (St. Louis, Missouri) and hypocretin 2exin B) was from American Peptide (Sunnyvale, California). Bothnsmitters were perfused at known concentrations through apcock assembly.

    hic factor Val66Met knock-inmice have both apical and basilar dendriticing atrophy of apical tuft dendrites from two-layer V pyramidal cells ingth of total apical dendrites (i.e., tuft oblique), apical tuft, and oblique1. (B) Basilar dendrites: (a) images illustrating basilar dendritic atrophy int/Met mice); (b) bar graph showing a significant decrease in total lengthtructure revealing a significant decrease in basilar dendrite intersectionsnosul(o...


View more >