Enhanced cognitive flexibility in reversal learning induced by removal of the extracellular matrix in auditory cortex
Edited by Michael Merzenich, Brain Plasticity Institute, San Francisco, CA, and approved January 3, 2014 (received for review May 31, 2013)
Significance
The brain’s extracellular matrix (ECM) mediates structural stability by enwrapping synaptic contacts fundamental for long-term memory storage. Whether the ECM in the adult brain might thereby govern learning-related plasticity, lifelong memory reformation, and higher cognitive functions is largely unknown. Here, we enzymatically degraded the ECM in the auditory cortex of adult Mongolian gerbils during a reversal learning task. Such local weakening of the structurally rigid ECM specifically accelerated strategy changes required for reversal learning. It neither affected general sensory learning nor erased already established, learned memory traces. Thus, ECM modulation might promote the cognitive flexibility that can build on learned behaviors, and has further implications to develop new tools for guided neuroplasticity with therapeutic potential.
Abstract
During brain maturation, the occurrence of the extracellular matrix (ECM) terminates juvenile plasticity by mediating structural stability. Interestingly, enzymatic removal of the ECM restores juvenile forms of plasticity, as for instance demonstrated by topographical reconnectivity in sensory pathways. However, to which degree the mature ECM is a compromise between stability and flexibility in the adult brain impacting synaptic plasticity as a fundamental basis for learning, lifelong memory formation, and higher cognitive functions is largely unknown. In this study, we removed the ECM in the auditory cortex of adult Mongolian gerbils during specific phases of cortex-dependent auditory relearning, which was induced by the contingency reversal of a frequency-modulated tone discrimination, a task requiring high behavioral flexibility. We found that ECM removal promoted a significant increase in relearning performance, without erasing already established—that is, learned—capacities when continuing discrimination training. The cognitive flexibility required for reversal learning of previously acquired behavioral habits, commonly understood to mainly rely on frontostriatal circuits, was enhanced by promoting synaptic plasticity via ECM removal within the sensory cortex. Our findings further suggest experimental modulation of the cortical ECM as a tool to open short-term windows of enhanced activity-dependent reorganization allowing for guided neuroplasticity.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Structural remodeling and stabilization of synaptic networks are key mechanisms underlying learning in the adult brain. During early life, high structural and functional plasticity is required for the experience-shaped development of basic neuronal circuits (1). With brain maturation, juvenile plasticity of so-called critical or sensitive periods is decreased and is accompanied by the appearance of the brain’s extracellular matrix (ECM) and its specialized compact form named “perineuronal net” (PNN) enwrappping cell bodies and synaptic contacts (2, 3). Enzymatic degradation of the ECM in adult animals has been demonstrated to restore such forms of developmental (juvenile) plasticity with respect to topographical map plasticity in the visual cortex (4), fear-response–mediating circuits in the amygdala (5), spinal cord injuries (6, 7), and song learning circuits of zebra finches (8). In addition, enzymatic ECM removal altered several forms of synaptic plasticity in vitro and in vivo (9–12). However, even though structural stability of networks acquired during developmental phases is essential for neuronal efficiency, mechanisms allowing synaptic remodeling are key events during learning and memory formation throughout life (13). We recently demonstrated that endogenous proteases moderately digesting specific components of the ECM are regulated in an activity-dependent manner (2, 14) and ECM removal modulates synaptic short-term plasticity by synaptic exchange of postsynaptic glutamate receptors (10, 15). Further, ECM modulation enhances synaptic short-term plasticity by affecting voltage-dependent calcium channels (9). These findings challenge the view of the purely stabilizing role of ECM in the brain and indicate a potential regulatory switch for plastic network adaptations within the adult brain at the level of individual synapses by modulating the extracellular space (16). However, it remains open to what extent ECM modulations influence learning-related plasticity in the adult brain and its specific effects on behavior during a cognitive task.
In the present study, we aimed at evaluating the potential role of experimental ECM removal within the auditory cortex (ACx) of Mongolian gerbils, which has been found to be particularly rich in ECM (17), during a cognitively demanding auditory go/no go shuttle-box task. We selected discrimination and reversal learning of frequency-modulated (FM) tones as a cognitive task, which necessarily requires ACx plasticity (18, 19). Injections of the ECM-degrading enzyme hyaluronidase (HYase) into the ACx after the first acquisition phase significantly enhanced subsequent reversal learning compared with sham-treated animals (injection of 0.9% saline). Particularly, after ECM degradation, animals abandoned the inappropriate discrimination strategy from the initial acquisition phase faster and thus promoted successful discrimination performance of the new contingency during reversal learning. ECM removal did not further influence the initial acquisition learning or interfere with already established—that is, learned—capacities in later learning stages, suggesting an enhanced activity-dependent neuroplastic reorganization of established synaptic networks in ACx during reversal learning.
Our findings suggest that experimental degradation of the ECM in sensory cortex, although not affecting general sensory learning, does, however, enhance the cognitive flexibility that can build on the learned behaviors. Thereby, ECM degradation could be used as a tool for guided neuroplasticity, which might also bear therapeutic potential.
Results
Specific forms of cortex-dependent auditory learning, as discrimination of FM tones (18), are ideal to test the functional role of the cortical ECM in ACx for adult learning capacities. In this study, we therefore used a cognitively demanding relearning task of FM discrimination with a contingency reversal between two FM tones. Mongolian gerbils (5–10 mo in age) were trained once per day in a two-compartmental go/no go shuttle-box paradigm to discriminate the modulation direction of two FM sounds (rising 2–4 kHz vs. falling 4–2 kHz). At a stable level of discrimination performance (7–9 d; phase 1), the contingency (i.e., go vs. no go) of both FM stimuli was reversed from one training session to the other (phase 2) in all reversal groups. Successful reversal learning could rely on (i) an alternative behavioral interpretation of the synaptic networks in ACx established during the learning task by higher processing areas (for instance, frontostriatal circuits) and/or (ii) a learning-dependent reorganization of the synaptic processing in ACx networks.
Enhanced Reversal Learning of FM Discrimination After ECM Removal.
First, we injected either the ECM-degrading enzyme HYase (group I, n = 11; 0.5 µL/500 units at three injection sites per hemisphere) or 0.9% saline (group II, n = 6; 0.5 µL at three injection sites per hemisphere) in bilateral ACx between the initial discrimination phase 1 and the relearning phase 2. In phase 1, hit rates showed a similar increase of discrimination performance across the first seven sessions of initial training in both groups. Hit rates increased across sessions to values above 0.5, whereas false-alarm rates stayed at values below 0.25, reflecting significant discrimination between conditioned stimuli (CS) CS+ and CS– for at least three consecutive sessions within phase 1 (Fig. 1 A and B, Upper). As a measure of discrimination performance, we further calculated d′ derived from signal detection theory to quantify the sensitivity of discrimination independent of experimental conditions biasing the response of an animal (Materials and Methods) (20). In both groups, d′ values showed a similar increase with sessions during the initial training phase 1, reaching values above 1 in the last three sessions of this phase. A two-factorial repeated-measures ANOVA (rmANOVA) on d′ values showed no significant differences with respect to the main factor “group” but significant differences for the main factor “session” (rmANOVA, α* = 0.05; Table S1, 1.1). Comparing the maximum d′ during the high level of performance within the last three sessions preceding the reversal, we found no significant differences for the factors group or “stable performance” (rmANOVA; Table S1, 1.2). We conclude that both groups showed no differences in the discrimination learning dynamics within phase 1.
Fig. 1.
After the initial training, animals underwent surgery and microinjections in ACx on 2 consecutive days (Materials and Methods). Group I was injected with HYase and group II with 0.9% saline. At 24 h after the second surgery, animals were trained with a reversed contingency compared with the initial acquisition phase 1. In the first session after the reversal, animals of both groups showed significantly higher false-alarm than hit rates, accounting for the conditioned rate (CR)+ responses to the CS–, which has been the CS+ in phase 1. Apparently, in the first reversal session, animals preserved the behavioral strategy based on the contingency learned during phase 1, which we therefore interpret as perseverative error. Accordingly, the respective d′ values in the first reversal session were negative (see Fig. 1, Bottom).
In the following reversal sessions, significant differences were observed in the relearning of the task between groups I and II. Comparison of d′ for all reversal sessions excluding the first session showed significant differences between the main factors of groups and sessions as well as their interaction (Table S1, 1.3). Also, the levels of d′ reached during the last three sessions of phase 2 were significantly different between groups (Table S1, 1.4). Animals injected with HYase reestablished a stable level of discrimination performance on the basis of the new contingency after the fourth session of phase 2. In group II, without HYase injection, this relearning was delayed by about six sessions, and animals reached highly significantly lower d′ levels at the end of phase 2. When comparing the stable levels of discrimination performance quantified by d′ in the last three sessions of learning phases 1 and 2, separately for each group, main factor phase was significantly different for group II, but not for group I (Table S1, 1.5 and 1.6). This indicates that the stable level of discrimination performance reached at the end of the initial phase and the relearning phase was similar only for the HYase-treated group. To account for effects in the individual subjects, each animal was classified as good relearner (reaching stable levels of discrimination performance comparable to initial training), as moderate relearner (showing at least three consecutive sessions with significant discrimination in phase 2), or as nonrelearner (only single, nonconsecutive sessions with a significant difference between hits and false alarms; Fig. S1). Out of 11 animals in group I, four animals were considered as good relearners, six as moderate relearners, and one as a nonrelearner. From the six animals in group II, only two were moderate relearners and four nonrelearners (Table S2). Henceforth, HYase treatment of ACx accelerated and strengthened the relearning of a contingency reversal of a cortex-dependent FM tone discrimination.
No Effect of ECM Removal on Initial Acquisition Learning.
To this point we have demonstrated enhanced auditory reversal learning by HYase injection in ACx. However, the initial FM discrimination also underlies cortical plasticity (18), and ECM removal could therefore enhance cortex-dependent learning also in naïve animals, which do not need to reconstitute an already learned behavioral strategy. Alternatively, activity-dependent modulation of neuronal networks immediately after the injection of HYase may lead to a stronger preservation of the initially learned behavioral strategy and thus to a further impairment of relearning. To assess such effects of ECM degradation during the acquisition phase, we trained group III (n = 6) animals in the acquisition phase 1 and the reversal phase 2, while injecting HYase already in naïve animals before the first acquisition training (Fig. 1C). We found the initial learning and levels of high, stable discrimination performance in phase 1 to be similar compared with both groups I and II (rmANOVAs of d′ showed no significant effects for groups or stable performance; Table S1, 1.1). However, relearning performance after the contingency change in group III mimicked results of group II after injection of 0.9% saline. Accordingly, rmANOVAs of d′ values in phase 2 between groups II and III did not show significant effects for the main factors of group and stable performance or their interaction (Table S1, 1.7). The enhanced relearning performance in group I thereby is reflected in significant effects compared with group III (rmANOVA, significant effects for group and stable performance; Table S1, 1.8). Individual relearning performance in group III was as in group II: two out of six animals were moderate relearners and four nonrelearners (Table S2). Thus, ECM degradation did not have an impact on initial discrimination learning of FM tones or the later relearning of discrimination after contingency reversal.
ECM Removal Does Not Interfere with Already Established Learning Capacities.
Next, we wondered whether the enhanced performance of group I in phase 2 can be relied on as an enhanced active relearning or is rather explained by a memory erasure of the initial discrimination learning. In the latter case, the increased reversal performance would reflect more another “naïve” training condition where the former training would not interfere with the reversed contingency. We trained nine animals (group IV) in FM discrimination that were injected after initial training with HYase in both auditory cortices as before in group I. However, in phase 2, those animals continued with the former discrimination task without a change of contingency. Animals showed no significant decrease in discrimination performance by HYase injection and stable discrimination for all consecutive sessions (Fig. 1D). Thus, HYase treatment of ACx did not erase the memory acquired in the initial learning phase (rmANOVA of d′ during the last three sessions of a phase; no significant effects for the main factors of phase and stable performance; Table S1, 1.9). Interestingly, comparison of the levels of high performance at the end of the phases between groups I and IV also revealed no significant difference for the main factors of phase or stable performance, indicating that animals of both groups, performing continued and reversed discrimination, finally performed discrimination of both FM tones equally well (Table S1, 1.10).
ECM Removal Impacts Correct Reassociation Speed by Immediate Effects on First Reversal Trials.
To reestablish a high and stable level of discrimination performance after contingency, reversal animals have to (i) abandon the initial discrimination strategy learned during the acquisition phase 1 and (ii) establish a new discrimination strategy based on the reversed contingency. Both behavioral processes could influence the observed significant differences in discrimination performance between our experimental groups. We therefore disentangled these processes by using an error-type analysis. The perseverative error in the first reversal session (d′ < 0) indicates that animals in all groups trained in the reversal task (Fig. 1 A–C) initially continued to perform the discrimination strategy learned during phase 1 at the beginning of phase 2. A more detailed analysis of the d′ block-wise over trials (Fig. 2A) within the first two reversal sessions showed significant differences between groups (rmANOVA; Table S1, 1.11). Perseveration of the now inappropriate discrimination strategy (d′ < 0) continued to exist for significantly longer durations in group II compared with group I (Fig. 2B). Elimination of perseverative errors (d′ ≥ 0) in group III required training durations that were intermediate between those found in groups I and II. Thus, ECM degradation immediately before contingency reversal in group I promoted quick rejection of the now inappropriate initial discrimination strategy.
Fig. 2.
To further assess the relearning performance, we quantified learning speed and asymptotic performance. To quantify learning speed, we determined in each animal the first training session before (phase 1) and after reversal (phase 2) in which a d′ ≥ 1 was found—that is, when this animal showed significant discrimination. Note that nonrelearners were also included in this analysis if this criterion was reached (for example, see Fig. S1C). To quantify asymptotic performance, the three highest d′ values found in phase 1 and phase 2 were separately averaged. In Fig. 3 A and B, the asymptotic performance was plotted against the learning speed for phase 1 and phase 2, respectively. In phase 1, no differences were observed between groups (Fig. 3A; rmANOVA; Table S1, 1.12 and 1.13). In phase 2, animals of group I showed significantly higher asymptotic performance (d′ = 1.60 ± 0.16) compared with groups II (d′ = 1.25 ± 0.25) and III (d′ = 1.27 ± 0.09). Also, animals in group I displayed a higher relearning speed as evidenced by reaching significant discrimination performance in an earlier session (d′ = 4.6 ± 0.48) compared with group II (d′ = 7.0 ± 1.0) and group III (d′ = 8.5 ± 1.3; rmANOVA; Table S1, 1.14). Please note that only a few animals of groups II and III showed at least one significant discrimination session during reversal learning, indicated by the lower number of data points in Fig. 3B.
Fig. 3.
Local Removal of the ECM in Adult ACx and Its Temporal Reappearance.
The ECM within ACx consists of compact PNNs enwrapping mainly parvalbumine-positive cells and the more diffuse ubiquitious ECM, both built up by proteoglycans like the chondroitin sulfate proteoglycans (CSPGs) and glycoproteins organized around the ECM’s backbone the glycosaminoglycan hyaluronic acid (2, 17). To specifically assess the depletion of the ECM within ACx, we injected the glycosidase HYase and 0.9% saline in opposing hemispheres of naïve animals at the age of ≥3 mo (n = 11), a time point when the mature ECM is formed (Fig. S2). HYase is digesting the glycosaminoglycan hyaluronic acid and with slower kinetics also the glycosaminglycan chondroitin sulfate (21). We injected HYase at three locations within the temporal cortex including primary and secondary auditory fields, as described earlier (22) (Fig. S3). Brains were removed and sliced at different days after the injection (3, 5, 7, 9, and 13 d) and stained with the well-established ECM marker Wisteria floribunda agglutinin (WFA), a lectin that specifically binds to chondroitin sulfates and thus labels CSPGs within the ECM (3, 4). WFA-based CSPG immunofluorescence was reduced mainly within the region of the temporal cortex containing primary and secondary auditory cortical fields (Fig. S3). WFA staining was considerably decreased 3 d after application of 500 units of HYase per injection site (Fig. 4A) at a level of around 33% compared with the control side at 3–7 d after the treatment (Kolmogorov–Smirnov test, **/***P < 0.01). Reconstitution of the ECM started 9 d after the injection, with WFA staining of 56.2 ± 12.8% compared with the control side (*P < 0.05). After 13 d, density of the ECM on the HYase-treated side was not significantly different from the control side (nonsignificant).
Fig. 4.
Discussion
Neuronal circuits formed by experience are stabilized during juvenile development to confer efficiency and permanency of basic neuronal processing, as for instance topographic connectivity in the sensory systems in mature animals (2, 4). This coincides with the appearance of the ECM, potentially limiting the flexibility of functional reorganization. However, during learning, synaptic networks are known to be reorganized (13), although the mechanisms of learning-related brain plasticity are still fairly unknown (23). In this study, we investigated the effects of ECM removal on a cognitively demanding behavioral task that is known to require learning-induced plastic reorganization of neuronal circuits in ACx (18, 19).
We used auditory reversal learning based on cortex-dependent FM tone discrimination in Mongolian gerbils while digesting the ECM in the ACx. Successful reversal learning requires inhibition of the initially learned behavioral strategy and, subsequently, relearning of the opposing behavioral strategy (24, 25). After acquiring robust discrimination of the FM modulation direction, untreated Mongolian gerbils failed or required a long period of training to behaviorally discriminate the two FM signals on the basis of the reversed contingency (group II; Fig. 1B). However, bilateral injection of HYase into ACx strongly promoted reversal learning (group I; Fig. 3). We found that ECM removal led to a quicker inhibition of the obsolete initial behavioral strategy (Fig. 2B). Further, ECM degradation accelerated relearning and enhanced the attained performance level of the opposing discrimination strategy (Fig. 3B). Moreover, we could demonstrate that initial HYase injection had no impact on discrimination learning during initial acquisition, nor significantly enhanced later relearning performance (group III; Fig. 1C and Fig. 3). Although, ECM removal did not interfere with or erase established memory traces: when initial training was continued after the injection, animals still successfully performed on the initial discrimination task (group IV; Fig. 1D).
The ECM was degraded significantly over 7–9 d after the treatment (Fig. 4B), so that the time window for activity-dependent changes extended presumably over several days, limited by subsequent ECM recovery. This is reflected by our finding that reversal learning was enhanced only in a time window of less than 9 d after HYase injection and hence apparently relied on a degraded ECM state. With ECM removal after contingency reversal in group I, this time window covered the whole reversal learning process including the behavioral inhibition of the obsolete discrimination strategy and the relearning of the opposing discrimination strategy, both of which were enhanced and occurred within about 4 d of training. When ECM was degraded at the beginning of the initial training (group III), only the inhibition of the obsolete discrimination strategy was slightly enhanced over day 9 and 10 after treatment—that is, when the ECM had not yet fully recovered. Thus, the observed behavioral effects were highly correlated with the current state of the ECM.
Potential Cellular Mechanisms of ECM-Derived Modulation of Neuroplasticity.
Synaptic function and plasticity can be altered by modulation of CSPGs and other ECM components—for instance, by leading to increased motility of postsynaptic spines (12). Recently, we demonstrated in vitro that a decrease of the perisynaptic ECM also modulates synapses at the functional level by enhancing lateral diffusion of glutamate receptors on the neuronal surface, which impacts synaptic short-term plasticity (10). Lack of several ECM proteins such as tenascin-R or brevican or chondroitin sulfates results in increased excitability of perisomatic interneurons, GABAA-dependent inhibition of long-term potentiation (LTP) induction (11, 26), impaired maintenance of LTP (27, 28), or impaired L-type voltage-dependent CA2+-channel (L-VDCC; Cav1.2)–dependent LTP induction (9). Thus, taken together, the ECM influences short- and long-term plasticity of synapses and thereby potentially regulates signal processing within a neuronal network during learning. Indeed, auditory associative learning has been correlated with structural changes of topographic connections within primary ACx (19, 29–31), changes in interneuron activity (32, 33), and Ca2+-dependent synaptic processing (34). However, further work will have to investigate the exact cellular and network effects of ECM removal in sensory cortex during learning processes in vivo.
Effects of ECM Removal on Learning Processes in Adults.
Growing evidence supports the view that dynamic ECM changes may be involved in providing important cues for synaptic stabilization or reorganization in the adult brain (2, 16). For instance, lack of components of the ECM in adult mice correlates with altered synaptic activity and plasticity (12, 14, 35). In our study, we have demonstrated that enzymatic ECM removal in ACx of adult Mongolian gerbils did not influence cortex-dependent learning of a new task or memory retrieval of an already established discrimination strategy. Our findings suggest that depletion of the ECM in the sensory cortex facilitated learning processes that require an activity-dependent modification of already existing synaptic networks without affecting already established cortex-based memories. Interference of memory traces thereby is a major problem in cortex-dependent tasks requiring cognitive flexibility like contingency reversal, set-, and task-shifting. In this respect, our findings support recent hypotheses emphasizing dynamic ECM changes in the adult brain to adeptly stabilize or free synaptic networks on a short time scale. Such processes might titrate long-term stability of memory content and its activity-dependent reorganization (2, 16, 22).
Necessarily, reversal learning requires plastic mechanisms involving different interacting networks compared with initial acquisition learning (24, 35, 36). In particular avoidance learning as investigated in our study depends on the formation of Pavlovian associations strongly influencing the instrumental avoidance responses. In reversal learning, the initially learned, rather inflexible Pavlovian associations must be inhibited, which could be achieved through a more flexible executive, goal-directed control of instrumental action. Interestingly, the brain circuits implied in reversal learning comprise the orbitofrontal cortex, the medial prefrontal cortex, and the medial striatum (36) and thus overlap with the circuits involved in goal-directed behavioral control (37) Although sensory cortex directly routes to frontal and striatal circuits (38, 39), the potential influences of sensory cortex on learning and cognitively flexible behavior are just starting to be understood (40). In this framework, our findings suggest that frontostriatal actions required for reversal learning and executive control might be recruited through an altered state of plasticity of sensory cortex triggered via ECM depletion. Thus, our findings reveal an important role in primary sensory cortex for learning, beyond a mere adaptation of the readout by subsequent processing stages (36, 41, 42).
Thus, ECM digestion might open a window of opportunities for learning-dependent plastic reorganizations, whereas the actual plastic changes are mediated by task-dependent interaction between sensory cortex and frontostriatal learning systems (43) and their neuromodulation (44). This interpretation would be in line with our finding of ECM degradation specifically enhancing cognitively more flexible forms of learning rather than general sensory learning itself.
Functional Implications.
On a more general account, our data show that the ECM might serve an important function for the regulation of stability and plasticity in learning-relevant synaptic networks. Recent findings implicate that structural changes in cortical circuits at the level of dendritic spines and individual synaptic rearrangements provide long-term information storage of prior experiences (12). On this microscopic level, the release or activation of ECM-removing or -modifying enzymes can indeed promote necessary neuronal adaptations to regulate learning-related plastic changes. This raises the interesting possibility that certain forms of native dynamic ECM turnover by endogenous proteolysis (14) might correlate with specific forms of cognitive abilities and learning capacities.
Finally, enzymatic ECM modifications have been implicated in therapeutic restoration processes after peripheral nerve damage—for example, in the spinal cord (6, 7). Our study now implicates that controlled ECM modulation might even bear potential to guide neuroplastic changes and support reprogramming and relearning of impaired cortical processing by improved experience-dependent fine-tuning without interfering with remaining functions. Invasive therapeutic approaches based on plastic reorganizations, as for instance in specific forms of deep brain stimulation (45) or auditory prostheses (20, 46), might, henceforth, include local activation or supply of ECM-modifying enzymes promoting synaptic rearrangements and stabilization for long-lasting successful treatment.
Materials and Methods
Behavioral Experiments and Intracortical Injections.
Adult gerbils (5–10 mo; n = 32) were trained once per day in a two-compartment shuttle box to discriminate the direction of linear FM modulations (250 ms tone with 5 ms linear onset and offset ramps, 250 ms pause, 4 s duration; rising, 2–4 kHz; falling, 4–2 kHz) as conditioned go/no go stimuli (CS+/CS–) in an active avoidance paradigm (18, 47). As unconditioned stimulus, foot shocks were administered through a metal floor grid individually adjusted for each animal (150–600 mA) to elicit comparable response strengths for the escape behavior (18, 20) (Fig. S4). Discrimination sensitivity was quantified by d′ values (20), with a d′ of 1.0 corresponding to a signal discrimination strength of one SD above noise.
Bilateral microinjections under anesthesia by i.p. injection of 0.4 mL/100 g body weight of 45% ketamine (50 mg/mL), 5% xylazine (2 mg/mL), and 50% isotonic 0.9% saline solution (154 mM) were performed at three locations covering the primary, anterior, and posterior fields of primary and secondary ACx (47). Per injection site, 500 nL of HYase solution (500 units) (groups I, III, and IV) or 0.9% saline (group II) were injected in 22 steps (22.8 nL each; 3 s pause; Nanoliter injector 2000, World Precision Instruments). For further information, see SI Materials and Methods.
Immunohistochemistry.
After unilateral HYase injections in naïve mature gerbils (≥3 mo; n = 11), the ECM was stained using WFA coupled with fluorescein isothiocyanate (Vector Laboratories) at several time points after the injections (3, 5, 7, 9, and 13 d). Fluorescence intensity was compared with the opposite hemisphere injected with 0.9% saline as a control. For detailed procedures, see SI Materials and Methods.
Acknowledgments
We thank Kathrin Ohl for technical assistance, Dr. Horst Schicknick for help with microinjection techniques, and Dr. Eike Budinger for assistance with anatomical analysis. The work was supported by grants from the Deutsche Forschungsgemeinschaft, the Schram Foundation (T287/21796/2011), and the Leibniz Institute for Neurobiology.
Supporting Information
Supporting Information (PDF)
Supporting Information
- Download
- 809.38 KB
References
1
TK Hensch, Critical period plasticity in local cortical circuits. Nat Rev Neurosci 6, 877–888 (2005).
2
ED Gundelfinger, R Frischknecht, D Choquet, M Heine, Converting juvenile into adult plasticity: A role for the brain’s extracellular matrix. Eur J Neurosci 31, 2156–2165 (2010).
3
MR Celio, I Blümcke, Perineuronal nets—A specialized form of extracellular matrix in the adult nervous system. Brain Res Brain Res Rev 19, 128–145 (1994).
4
T Pizzorusso, et al., Reactivation of ocular dominance plasticity in the adult visual cortex. Science 298, 1248–1251 (2002).
5
N Gogolla, P Caroni, A Lüthi, C Herry, Perineuronal nets protect fear memories from erasure. Science 325, 1258–1261 (2009).
6
CL Tan, et al., Kindlin-1 enhances axon growth on inhibitory chondroitin sulfate proteoglycans and promotes sensory axon regeneration. J Neurosci 32, 7325–7335 (2012).
7
C Bowes, JM Massey, M Burish, CM Cerkevich, JH Kaas, Chondroitinase ABC promotes selective reactivation of somatosensory cortex in squirrel monkeys after a cervical dorsal column lesion. Proc Natl Acad Sci USA 109, 2595–2600 (2012).
8
TS Balmer, VM Carels, JL Frisch, TA Nick, Modulation of perineuronal nets and parvalbumin with developmental song learning. J Neurosci 29, 12878–12885 (2009).
9
G Kochlamazashvili, et al., The extracellular matrix molecule hyaluronic acid regulates hippocampal synaptic plasticity by modulating postsynaptic L-type Ca(2+) channels. Neuron 67, 116–128 (2010).
10
R Frischknecht, et al., Brain extracellular matrix affects AMPA receptor lateral mobility and short-term synaptic plasticity. Nat Neurosci 12, 897–904 (2009).
11
O Bukalo, M Schachner, A Dityatev, Modification of extracellular matrix by enzymatic removal of chondroitin sulfate and by lack of tenascin-R differentially affects several forms of synaptic plasticity in the hippocampus. Neuroscience 104, 359–369 (2001).
12
L de Vivo, et al., Extracellular matrix inhibits structural and functional plasticity of dendritic spines in the adult visual cortex. Nature communications 4, 1484 (2013).
13
SB Hofer, TD Mrsic-Flogel, T Bonhoeffer, M Hübener, Experience leaves a lasting structural trace in cortical circuits. Nature 457, 313–317 (2009).
14
R Frischknecht, A Fejtova, M Viesti, A Stephan, P Sonderegger, Activity-induced synaptic capture and exocytosis of the neuronal serine protease neurotrypsin. J Neurosci 28, 1568–1579 (2008).
15
M Heine, et al., Surface mobility of postsynaptic AMPARs tunes synaptic transmission. Science 320, 201–205 (2008).
16
A Dityatev, M Schachner, P Sonderegger, The dual role of the extracellular matrix in synaptic plasticity and homeostasis. Nat Rev Neurosci 11, 735–746 (2010).
17
G Brückner, et al., Cortical areas are revealed by distribution patterns of proteoglycan components and parvalbumin in the Mongolian gerbil and rat. Brain Res 658, 67–86 (1994).
18
FW Ohl, W Wetzel, T Wagner, A Rech, H Scheich, Bilateral ablation of auditory cortex in Mongolian gerbil affects discrimination of frequency modulated tones but not of pure tones. Learn Mem 6, 347–362 (1999).
19
FW Ohl, H Scheich, Learning-induced plasticity in animal and human auditory cortex. Curr Opin Neurobiol 15, 470–477 (2005).
20
M Deliano, H Scheich, FW Ohl, Auditory cortical activity after intracortical microstimulation and its role for sensory processing and learning. J Neurosci 29, 15898–15909 (2009).
21
R Stern, MJ Jedrzejas, Hyaluronidases: Their genomics, structures, and mechanisms of action. Chem Rev 106, 818–839 (2006).
22
M Kraus, et al., Memory consolidation for the discrimination of frequency-modulated tones in mongolian gerbils is sensitive to protein-synthesis inhibitors applied to the auditory cortex. Learn Mem 9, 293–303 (2002).
23
Y Dudai, The restless engram: Consolidations never end. Annu Rev Neurosci 35, 227–247 (2012).
24
R Dias, TW Robbins, AC Roberts, Dissociable forms of inhibitory control within prefrontal cortex with an analog of the Wisconsin Card Sort Test: Restriction to novel situations and independence from “on-line” processing. J Neurosci 17, 9285–9297 (1997).
25
DJ Watson, ME Stanton, Spatial discrimination reversal learning in weanling rats is impaired by striatal administration of an NMDA-receptor antagonist. Learn Mem 16, 564–572 (2009).
26
S Miyata, A Akagi, N Hayashi, K Watanabe, A Oohira, Activity-dependent regulation of a chondroitin sulfate proteoglycan 6B4 phosphacan/RPTPbeta in the hypothalamic supraoptic nucleus. Brain Res 1017, 163–171 (2004).
27
XH Zhou, et al., Neurocan is dispensable for brain development. Mol Cell Biol 21, 5970–5978 (2001).
28
C Brakebusch, et al., Brevican-deficient mice display impaired hippocampal CA1 long-term potentiation but show no obvious deficits in learning and memory. Mol Cell Biol 22, 7417–7427 (2002).
29
JM Edeline, NM Weinberger, Receptive field plasticity in the auditory cortex during frequency discrimination training: Selective retuning independent of task difficulty. Behav Neurosci 107, 82–103 (1993).
30
GH Recanzone, CE Schreiner, MM Merzenich, Plasticity in the frequency representation of primary auditory cortex following discrimination training in adult owl monkeys. J Neurosci 13, 87–103 (1993).
31
MD Keeling, BM Calhoun, K Krüger, DB Polley, CE Schreiner, Spectral integration plasticity in cat auditory cortex induced by perceptual training. Exp Brain Res 184, 493–509 (2008).
32
EE Galindo-Leon, FG Lin, RC Liu, Inhibitory plasticity in a lateral band improves cortical detection of natural vocalizations. Neuron 62, 705–716 (2009).
33
JJ Letzkus, et al., A disinhibitory microcircuit for associative fear learning in the auditory cortex. Nature 480, 331–335 (2011).
34
H Salgado, et al., Muscarinic M2 and M1 receptors reduce GABA release by Ca2+ channel modulation through activation of PI3K/Ca2+-independent and PLC/Ca2+-dependent PKC. J Neurophysiol 98, 952–965 (2007).
35
A Dityatev, T Fellin, Extracellular matrix in plasticity and epileptogenesis. Neuron Glia Biol 4, 235–247 (2008).
36
AA Kehagia, GK Murray, TW Robbins, Learning and cognitive flexibility: Frontostriatal function and monoaminergic modulation. Curr Opin Neurobiol 20, 199–204 (2010).
37
P Dayan, ND Daw, Decision theory, reinforcement learning, and the brain. Cogn Affect Behav Neurosci 8, 429–453 (2008).
38
BW Balleine, JP O’Doherty, Human and rodent homologies in action control: Corticostriatal determinants of goal-directed and habitual action. Neuropsychopharmacology 35, 48–69 (2010).
39
F Bordi, J LeDoux, Sensory tuning beyond the sensory system: An initial analysis of auditory response properties of neurons in the lateral amygdaloid nucleus and overlying areas of the striatum. J Neurosci 12, 2493–2503 (1992).
40
E Budinger, P Heil, H Scheich, Functional organization of auditory cortex in the Mongolian gerbil (Meriones unguiculatus). III. Anatomical subdivisions and corticocortical connections. Eur J Neurosci 12, 2425–2451 (2000).
41
KK Chadman, DJ Watson, ME Stanton, NMDA receptor antagonism impairs reversal learning in developing rats. Behav Neurosci 120, 1071–1083 (2006).
42
D Schiller, MR Delgado, Overlapping neural systems mediating extinction, reversal and regulation of fear. Trends Cogn Sci 14, 268–276 (2010).
43
P Znamenskiy, AM Zador, Corticostriatal neurons in auditory cortex drive decisions during auditory discrimination. Nature 497, 482–485 (2013).
44
ME van der Schaaf, et al., Establishing the dopamine dependency of human striatal signals during reward and punishment reversal learning. Cereb Cortex, 2012).
45
M Heldmann, et al., Deep brain stimulation of nucleus accumbens region in alcoholism affects reward processing. PLoS ONE 7, e36572 (2012).
46
HH Lim, et al., Electrical stimulation of the midbrain for hearing restoration: Insight into the functional organization of the human central auditory system. J Neurosci 27, 13541–13551 (2007).
47
H Schicknick, et al., Dopaminergic modulation of auditory cortex-dependent memory consolidation through mTOR. Cereb Cortex 18, 2646–2658 (2008).
Information & Authors
Information
Published in
Classifications
Copyright
Freely available online through the PNAS open access option.
Submission history
Published online: February 3, 2014
Published in issue: February 18, 2014
Keywords
Acknowledgments
We thank Kathrin Ohl for technical assistance, Dr. Horst Schicknick for help with microinjection techniques, and Dr. Eike Budinger for assistance with anatomical analysis. The work was supported by grants from the Deutsche Forschungsgemeinschaft, the Schram Foundation (T287/21796/2011), and the Leibniz Institute for Neurobiology.
Notes
This article is a PNAS Direct Submission.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Altmetrics
Citations
Cite this article
Enhanced cognitive flexibility in reversal learning induced by removal of the extracellular matrix in auditory cortex, Proc. Natl. Acad. Sci. U.S.A.
111 (7) 2800-2805,
https://doi.org/10.1073/pnas.1310272111
(2014).
Copied!
Copying failed.
Export the article citation data by selecting a format from the list below and clicking Export.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.