Approach/ Methods
Animals: For all experiments, adult male Sprague Dawley rats (Harlan, 249-279g) will be used. Rats will be group housed upon arrival, after a period of 1-week acclimation, they will be single-housed and provided with food and water ad libitum. Experiments will take place during the light cycle (8:00-17:00hr). For all aims, group sizes were determined to be n=12 rats/per group; this was detected using power analysis by predicting a mean difference of 35%, with standard deviation of 25%, with significance detected at p<0.05 by a two-tailed test with power=0.80.
Chronic Unpredictable Stress (CUS): Chronic unpredictable stress will occur as previously described.[15] CUS consists of 14 days of alternating stressors. The stressors are: 30-minute restraint, 1 hour shake, 45 minutes of social defeat, 10 minutes of tail pinch, 24 hr of wet bedding, 15 minutes of mild footshock, and swim stress. Procedurally, restraint entails placing the rat in a nylon tube with velcro straps. For social defeat, Spragues are pinned down by a Long Evans retired breeder. Tail pinch stress involved placing the rat in a restraint tube with a clothespin 1 cm from the base of the tail. Wet bedding consists of saturating the rat’s cage bedding with 800 ml of water. Shake stress involves placing 6 Sprague rats on a shaker that generates 220 movements/min for 1 hr. Lastly, mild footshock involves placing a rat in a chamber with a grid floor attached to a shock box, the rat receives a shock of 1.5 mA every 3 minutes at 30 second intervals for 5 seconds, for a total of 15 minutes.
Fear Extinction: Two days prior to the start of CUS procedures, rats are habituated to two contexts in sound attenuating cabinets for 15 minutes each. Context A consists of a metal conditioning chamber (30.5 x 25.4 x 30.5 cm; model H10-11RC-TC-Sf) with a grid floor attached to a shock box (model H13-15). Context B is not associated with a shock and is a different chamber with a green and white vinyl floor and white vinyl walls.
Day 0—Fear Conditioning: Rats receive fear conditioning or tone control treatment in Context A the day before the start of CUS. Fear conditioning consists of 4 pairings of a tone (10 kHz, 75 dB, 20 s) with a footshock (0.8 mA, 0.5 s), with an inter-trial interval of 120 s between tones. Freezing during each tone will be quantified videographically by FreezeView software (ActiMetrics #ACT-100, Coulbourn Instruments. Freezing is defined as movement below the motion index threshold of 10 for at least 1 second. Rats will be assigned to groups so that the percent of time freezing on tone 4 is comparable between groups. Rats that will be exposed to the tones but do not receive a shock pairing are referred to as tone controls.
Day 17— Fear Extinction: Three days after the end of CUS, rats will receive a single fear extinction treatment in Context B. During fear extinction, rats are exposed to the tone alone for 16 trials, with an inter-trial interval of 120 s. On average, rats that have been fear conditioned will exhibit 70% freezing at the first tone, this decreases to about 25% freezing on the last tone.
Immunohistochemistry: On day 17, 1 hr after completing extinction, rats will be anesthetized and perfused transcardially with 4% paraformaldehyde. The PFC will be sectioned at 40 μm, and immunohistochemistry will be performed on free-floating slices. To quench peroxidases, slices will be incubated in 3% hydrogen peroxide in phosphate-buffered saline. For immunohistochemistry detection, slices will be incubated in rabbit polyclonal antibodies for ΔFosB (SC-48, 1:200; Santa Cruz Biotechnology), pS6-240/244 (1:20,000; Cell Signaling, Beverly, MA), CamKII (1:10,000; Invitrogen, Waltham, MA). Then, they will be incubated in HRP-conjugated anti-mouse and rabbit secondary antibody (1:2,000; Cell Signaling, Beverly, MA), and then in cyanine 3-tagged tyramide reagent. Slices will be viewed using a Nikon Microphot-SA microscope (confocal specs?). The cells that show colocalized expression of pS6, FosB and CamKII will be counted for three sections per rat in a 200 μm2 field of view.
Stereotaxic surgery for viral delivery and microinjections: For Aim 1b, animals will receive viral delivery of 0.5 μl of AAV2-hSyn-eNpHR3.0-EYFP (or the control virus AAV2-hsyn-EYFP) bilaterally, and an optic cannulae implanted into the vmPFC (AP +2.9, ML +1.0, DV -4.1). The virus will be allowed to express for six weeks, and then undergo fear conditioning or tone control treatment on day 0. Rats will receive CUS or control treatment from days 1-14, and remain undisturbed from days 15-16. On Day 17, for the extent of extinction treatment, optical fibers will be connected to a 100-mW 593nm DPSS Laser to deliver yellow light continuously for 32 minutes. For Aims 2a-2c, rats will be implanted with cannula (Plastics One, Roanoke, VA) at the level of the vmPFC at an 11 angle. Animals will be allowed to recover for 1 week after the surgery. On day 0, they will receive fear conditioning or tone exposure, on days 1-14 they will undergo CUS. On days 15-16, they will remain undisturbed. On day 17, they will receive a microinjection prior to extinction training. Rats will receive either a vehicle microinjection or sheep anti-BDNF (EMD Millipore, Billerica, MA) at a dose of 0.5mg/0.5ml at a rate of 0.1 ul/min per side. The injector will remain in place for 2 min to allow for diffusion. Twenty minutes after the completion of the microinjections, the rats will undergo extinction.
Electrophysiology: Rats will be anesthetized with chloral hydrate (400mg/kg, i.p.) and placed in a stereotaxic apparatus. To record evoked local field potentials in the vmPFC, a bipolar concentric stimulating electrode with a tip extending 1mm (Plastics One, Roanoke, VA) will be placed in the MDT (AP: -2.5, ML: 0.9, DV: -5.0 mm), and a tungsten recording electrode positioned at the ventromedial prefrontal cortex (-3.0 AP; +0.6 ML; -4.0 DV). Local field potentials in the vmPFC will be generated by stimulating the MDT (30 pulses, 260 s pulse width, 0.1 Hz) at 100 uA increments. The recorded potentials will then be amplified and filtered (gain 5000x; 0.3-1000 Hz) and digitized (Power Lab, ADInstruments) at a 2 kHz sampling rate for offline analysis. The quantified response will be measured as the amplitude between the peak of the first negative deflection and the peak of the subsequent positive deflection.[16]
vmPFC dissections and Western Blotting: Rats will be killed via rapid decapitation. Using a brain matrix on ice, the vmPFC will be dissected from a 2mm coronal slab which will be cut 2-4 mm caudal to the frontal pole. The cortex tissue adjacent and medial to the internal capsule will be dissected from this slab and rapidly frozen in isopentane. Tissue will be stored at -80 C until use.
Tissue samples will be sonicated (3 x 5s, at 50% power) in RIPA buffer (50nM Tris-HCl pH 8, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS; Halt protease and phosphatase inhibitor cocktail [Thermo Scientific, Rockford, IL]). After sonication, the samples with RIPA will be left to incubate on ice for 30 min and then will be centrifuged at 16,000 xg for 25 mins at 4 C. The supernatants will then be transferred to fresh tubes, and protein concentration will be determined for each sample using the Bradford assay (Sigma, St. Louis, MO). Thirty micrograms of total sample protein will be loaded on a 4-12% NuPage Bis-Tris SDS gels (Invitrogen) and electrophoresed at 100 V for an hour. Subsequently, proteins will be transferred to PVDF using the iBlot transblot system (Invitrogen). Membranes will then be incubated with polyclonal antibodies to pS6-S240/244 (1:2,000, Cell Signaling) and GAPDH (1:20,000; Cell Signaling) to normalize for loading, and will then be incubated in anti-rabbit secondary antibody (1:10,000) and will be detected using ECL Prime detection agent (GE Healthcare, Little Chalfont, UK). The membranes will then be stripped to be re-probed with anti-S6 antibody (1:1,000, Cell Signaling) and then with secondary anti-mouse antibody (1:10,000; Cell Signaling).
Attentional Set Shifting Task (AST)
To measure cognitive flexibility, we will measure performance on the extradimensional set shifting stage of the attentional set shifting test (AST). One day after the last day of CUS, rats will be taken through behavioral procedures for the AST as previously described.[13] The animals will be tested in a white wooden box (75 x 44 x 30 cm) containing a start gate at one end and two terracotta pots (diameter 7 cm, depth 6 cm) separated by a Plexiglas wall at the other end.
i. Habituation day: Rats will be trained to dig in sawdust to retrieve a food reward, half a Honey Nut Cheerio (General Mills Cereals, Minneapolis, MN, USA).
ii. Training day: Rats will be trained to locate the reward in one of the pots by discriminating cues in two stimulus dimensions, an odor applied to the rim of the pot, and the digging medium that filled the pot.
iii. Treatment day: One day after training day, animals will receive a microinjection of anti-BDNF in the vmPFC prior to extinction or vehicle, or an infusion of BDNF without extinction treatment.
iv. Testing day on AST: The rats will be tested on a series of discrimination tasks, in which a criterion of six consecutive correct responses is required to proceed to the next task. Rats will be tested using either medium or odor as the relevant cue in the early discrimination tasks (simple discrimination, complex discrimination, reversal learning task, intradimensional shift, second reversal). In the extra-dimensional (ED) set shifting task, the previously relevant dimension (such as odor), is irrelevant while the other dimension (medium) is predictive of the cheerio reward location. To account for differences in medium-odor, odor-medium performance, half of the rats will be tested with odor being the initial relevant stimulus, and the other half will be tested with medium as the relevant stimulus. The dependent measure will be the number of trials required to meet criterion. Animals that do not dig within 10 min on six consecutive trials, or do not reach criterion within 50 trials will be excluded from analysis.
Experimental Design:
Specific Aim 1— Aim 1. Investigate FE-induced protein translation in glutamatergic neurons of the vmPFC.
Aim 1a. Investigate whether FE is inducing activity-dependent protein translation in glutamatergic vmPFC neurons that have been activated by CUS.
Rationale: Four observations support the rationale for this subaim: 1) Stress literature suggests that glutamatergic neurons in the PFC are susceptible to stress-induced damage, presumably through their repeated activation during chronic stress; 2) silencing glutamatergic neurons blocks the therapeutic effects of FE on cognitive flexibility,[15] 3) activity-dependent protein synthesis is necessary for the effects of FE in the vmPFC of stressed animals, and lastly, 4) pS6, a marker for protein synthesis, is only induced by FE in the vmPFC of stressed animals.[13] Therefore, I hypothesize that glutamatergic neurons are repeatedly activated by stress, and FE induced protein translation in these neurons. To test this hypothesis, I will subject rats to stress and FE and perform triple labeling of ΔFosB, CamKII, and pS6. Rats will be assigned to one of four groups defined by Stress X Extinction. On day 0, rats will undergo either fear conditioning or tone control treatment. On Days 1-14, rats will undergo either CUS or daily handling procedures. To mimic the timeline of AST (habituation and training on days 15-16), rats will be left undisturbed from days 15-16. On day 17, rats will undergo fear extinction treatment. One hour after the end of FE, rats will be transcardially perfused with 4% PFA and their brains collected for immunohistochemistry procedures. For IHC, sections will be cut at 40 um on a cryostat. After IHC detection for ΔFosB, CamKII, and pS6, I will quantify the number of positive triple-labeled cells (3 slices per rat; 200 um2 field view). I will analyze the data using a 2-way ANOVA (Stress X Extinction), with significance set to p>0.05. I hypothesize that the stress/fear extinction group will have a greater number of triple-labeled cells compared to the remaining groups. To compare measures between groups, I will use a post-hoc analysis (Neuman Kewls), with significance set at p>0.05.
Expected Results and Alternatives: I expect that FE after CUS will increase the number of cells that co-express CamKII, ΔFosB, and pS6. This result would suggest that FE induces protein translation in pyramidal cells (CamKII+ neurons) that are activated by chronic stress. Alternatively, negative results could show that FE induces pS6 in glutamatergic cells that were not activated stress and hence, do not colocalize with ΔFosB. Since GABAergic interneurons can express both pS6 and ΔFosB, it is plausible that pS6 is induced by extinction in these cells. To test this possibility, sections from the same animals will be taken and probed for anti-GAD67 to label GABAergic cells. However, I do not predict this to be the case since previous experiments show that after silencing glutamatergic neurons after extinction, c-fos activation in the vmPFC is scarce. For the second scenario, FE may induce pS6 in CamKII cells that were not activated by stress, suggesting that FE induces protein synthesis in “non-stressed” glutamatergic neurons, thereby providing a novel observation for future testing.
Aim 1b) Determine whether FE-induced protein translation in stressed animals is driven by input from the ventral hippocampus to the vmPFC.
Rationale: We have shown that silencing glutamatergic neurons in the vmPFC using Gi-DREADDs during FE blocks the effects of FE on set shifting in stressed animals. Moreover, briefly activating glutamatergic neurons in the vmPFC using Gq-DREADDs after CUS is sufficient to reverse deficits in set shifting.[15] While these observations point to the importance of pyramidal cell activity for the effects of extinction, it is unknown whether plasticity in the vmPFC is afferent-driven. The vHipp sends projections to the vmPFC that are vital to extinction learning, and the vHipp can modulate vmPFC excitability. Further, the vHipp releases BDNF during extinction that facilitates extinction learning. [17] Therefore, the vHipp is a likely effector of protein translation in the IL during FE. I hypothesize that silencing the vHipp-vmPFC projection pathway will inhibit the induction of pS6 in glutamatergic vmPFC neurons after extinction. To test this hypothesis, I perform stereotaxic surgery to inject an adeno-associated virus (AAV) virus carrying either an inhibitory opsin (Archeordopsin) or enhanced yellow fluorescent protein (eYFP) under the control of a CamKII promoter, into either the ventral hippocampus (coordinates) or the medial dorsal thalamus (another afferent coordinates), and an optical fiber implanted in the vmPFC (coordinates). The rats will remain undisturbed for 6 weeks to allow the virus to express. After that period, rats will undergo FC on Day 0, CUS on days 1-14, a rest period from days 15-16. On day 17, the animals will receive green light (532mm) in the vmPFC for the duration of FE treatment. One hour after the end of FE, rats will be perfused to fix brains. mPFC sections will be collected and stained using IHC for anti-YFP, anti-Arche, CamKII, and pS6.
Expected Results and Alternatives: A positive result would suggest that activity from the vHipp is driving plasticity in glutamatergic neurons of the vmPFC during FE. This finding would provide a mechanism through which FE induces plasticity in the vmPFC. Future studies could investigate whether inhibiting this pathway blocks the effects of FE on evoked LFPs, as well as set shifting, after chronic stress. A negative result would suggest that input from the vHipp is not necessary for protein translation occurring in the vmPFC; this result could indicate that another afferent is promoting plasticity in the vmPFC or that activity in the vmPFC itself is exerting homosynaptic plasticity.
Specific Aim 2— Determine the role of BDNF and TrkB activation in measures of plasticity after CUS in the vmPFC.
Aim 2a) Determine if TrkB activation is necessary for pS6 induction in the vmPFC after CUS and FE.
Rationale: Several observations support the importance of BDNF for fear extinction in the vmPFC. Individuals with a polymorphism that causes decreased BDNF release show impaired extinction learning, and blocking BDNF in the vmPFC within extinction sessions impairs extinction learning.[11, 17] FE increases BDNF mRNA in the vmPFC, and preliminary data shows that FE induces TrkB phosphorylation[18]. FE activates the TrkB receptor (Fig 3?) resulting in the phosphorylation of the Y515 site, which initiates the PI3K-AKT pathway, a pathway that regulates ribosomal protein S6 kinase, and thus phosphorylates ribosomal protein S6. Therefore, FE likely activates TrkB through PI3K-AKT and, in doing so, regulates pS6. However, since other factors may regulate pS6, it is unknown whether BDNF-mediated activation of TrkB is necessary for the induction of pS6. I hypothesize that BDNF is necessary for the induction of pS6 after FE in the vmPFC of stressed animals. To test whether sequestering BDNF during extinction blocks the induction of pS6, I will implant a guide cannula in the vmPFC at a 11 angle and allow the animals to recover for a week. I will have a total of ___ , with a total of 8 groups, as defined by injection (vehicle or anti-BDNF) X (stress/control) X (fear extinction/tone controls). On Day 0, they will receive FC or tone control treatment. From Days 1-14, they will undergo CUS. On days 15-16, the rats will be left undisturbed. On day 17, rats will receive a microinjection of either vehicle or anti-BDNF into the vmPFC. 30 minutes after the microinjection, rats will undergo FE. 1 hr after FE, rats will be rapidly decapitated and their vmPFC dissected for western blot analysis with antibodies against phosphorylated S6 (Cell Signaling) and pan-S6 (Cell Signaling) and GAPDH (Cell Signaling). Data will be expressed as a ratio pS6/total S6/GAPDH? Data will be analyzed by multi-factorial ANOVA (injection x FE x stress), and post hoc comparisons will be analyzed using Newman Keuls with significance set at p<0.05.
Expected results and alternative outcomes. I expect that blockade of BDNF during FE will result in the reduction of FE-induced pS6 in the vmPFC of stressed animals. This result would suggest that BDNF is necessary to induce de novo protein translation in the vmPFC of stressed animals. A negative result would suggest that other factors may be inducing pS6, independent of TrkB activation. In this case, I would probe for …
Aim 2b) Investigate if TrkB activation is necessary and sufficient for the effects of FE on cognitive set shifting.
Rationale: FE results in an increase of BDNF in the vHPC that increases the firing rate of pyramidal neurons in the vmPFC and enhances extinction learning. [17] Further, the induction of extinction by infusing BDNF in the vHPC can be inhibited by administering anti-BDNF in the vmPFC.
Several groups have shown that BDNF is involved in the effects of antidepressants through the activation of TrkB. Additionally, Quirk and colleagues have shown that exogenously administered BDNF in the vmPFC enhances fear extinction behavior and is sufficient to induce extinction. Further, BDNF infusion into the vmPFC mimics the effects of FE on the firing properties of glutamatergic neurons in the vmPFC.
NEED MORE Thus, it is possible that infusing BDNF into the vmPFC is sufficient and necessary to induce plasticity that restores vmPFC function. Indeed, BDNF is involved in the regulation of several factors that could enhance mPFC function, such as increasing AMPAR trafficking to the postsynaptic density and enhancing glutamate release. I have collected some preliminary data that suggests that stress reduces surface GluR1, which is brought back to control levels by FE. I hypothesize that blocking BDNF during FE in stressed animals will reverse the effects of FE on set shifting, and that infusion of BDNF into the vmPFC is sufficient to reverse stress-induced deficits on set-shifting. To test the necessity of BDNF, animals will undergo cannula implantation surgery in the vmPFC at an angle of 11 (similar as above). On day 0, they will receive FC or tone treatment, on days 1-14 they will undergo CUS or control treatment. On days 15-16, animals will be trained on AST. On day 17, rats will receive a microinjection infusion of anti-BDNF or vehicle into the vmPFC 30 mins prior to FE treatment. On day 18, rats will be tested on set shifting on the AST. To test whether BDNF alone reverses CUS-induced deficits on set shifting, rats will undergo the same timeline as described above from days 0-16. On day 17, rats will receive an infusion of human recombinant BDNF protein (0.75 ug/side) into the vmPFC. On day 18, they will be tested on the AST.
Expected results and outcomes: If blocking BDNF in the vmPFC prevents FE from reversing CUS-induced deficits on set shifting, this result would suggest that BDNF is inducing protein translation during FE that is necessary for the therapeutic effects of FE. Since BDNF is necessary for FE consolidation, a negative result BDNF-mediated mechanisms that regulate consolidation after FE and those that regulate set-shifting are independent.
Aim 2c) Test whether TrkB activation in the vmPFC during FE is necessary for its effects on vmPFC responsivity to excitatory afferent input.
Rationale: Chronic stress causes a decrease in spike firing frequency in pyramidal cells of the vmPFC.[19] We have shown that chronic stress causes a decrease in evoked LFPs in the vmPFC that is reversed by FE.[15] Since BDNF enhances glutamate release in the mPFC and regulates AMPA receptor trafficking to the postsynaptic density, it is possible that BDNF is mediating the restorative effect of FE on compromised responsivity of the vMPFC. A potential mechanism through which FE might be restoring vmPFC is by activating the BDNF/TrkB signaling pathway and therein causing plastic changes in the vmPFC. To test the hypothesis that TrkB activation is necessary for the effects of FE on evoked LFPs, I will block BDNF signaling in the vmPFC during FE and test the effects on evoked LFPs in the vmPFC 24 hrs after FE—terrible sentence. Animals will be implanted with a cannula in the vmPFC at an angle of 11. On day 0, they will receive FC or tone treatment, on days 1-14 they will undergo CUS or control treatment. On days 15-16, animals remain undisturbed. On day 17, rats will receive a microinjection of anti-BDNF or vehicle into the vmPFC 30 mins prior to FE treatment. On day 18, evoked LFPs will be recorded by stimulating the MDT and recording in the vmPFC. The response to stimulation will be quantified as the amplitude between the N1 and P2 complex [16], and will be analyzed using a 2-way ANOVA with four groups (defined by stress x extinction). Post hoc comparisons will be analyzed with Neuman Kewls pairwise comparisons, with significance set at p<0.05.
Expected results and outcomes: A positive result would show that BDNF during FE in stressed animals is promoting plasticity that reverses CUS-induced population responses in the vmPFC. BDNF could likely be doing so through several mechanisms— such as increasing AMPAr trafficking to the postsynaptic density, facilitating the release of glutamate at synaptic terminals, and/or increasing dendritic complexity and dendritic spine density. Given positive results, I will investigate the mechanisms through which BDNF might be increasing restoring responsivity in the vmPFC. Conversely, a negative result would suggest that FE does not require BDNF to promote plasticity that results in restoring responsivity in the stressed vmPFC. This result would be intriguing as well, and could suggest that other growth factors involved in protein translation during FE are driving plasticity in the vmPFC.