Npgrj_nn_2271 256.258

In this human study, we tested the hypotheses that the fear response can be weakened by disrupting the reconsolidation process and that disrupting the reconsolidation of the fear memory will prevent the return of fear. To test these hypotheses, we used a differential fear-conditioning procedure with fear-relevant stimuli. Testing includeddifferent phases across 3 d: fear acquisition (day 1), memory reactiva- Merel Kindt, Marieke Soeter & Bram Vervliet tion (day 2), and extinction followed by a reinstatement procedure anda test phase (day 3) (Supplementary Figs. 1 and 2 online). The Animal studies have shown that fear memories can change conditioned fear response was measured as potentiation of the eyeblink when recalled, a process referred to as reconsolidation. We startle reflex to a loud noise (40 ms, 104 dB) by electromyography of the found that oral administration of the b-adrenergic receptor right orbicularis oculi muscle. Stronger startle responses to the loud antagonist propranolol before memory reactivation in humans noise during the fear-conditioned stimulus (CS1+) as compared with erased the behavioral expression of the fear memory 24 h later the control stimulus (CS2–) reflects the fearful state of the participant and prevented the return of fear. Disrupting the reconsolidation elicited by CS1+. Startle potentiation taps directly into the amygdala, of fear memory opens up new avenues for providing a long-term and fear-conditioning procedures yield highly reliable and robust cure for patients with emotional disorders.
startle potentiation. In addition, declarative knowledge of the con-tingency between the conditioned stimulus and the unconditioned Since the dawn of psychology at the end of the nineteenth century, stimulus was measured through online shock-expectancy ratings All rights reser
psychologists and psychiatrists have tried with dozens of pharmacolo- during each conditioned stimulus presentation. Reconsolidation of gical and psychological treatments to change undesired emotional fear memory was manipulated by administration of propranolol memory. However, even the most effective treatments only eliminate (40 mg, n ¼ 20), randomized and double-blind placebo controlled fearful responding, leaving the original fear memory intaas is (n ¼ 20) (see Supplementary Methods online). For the additional substantiated by the high percentages of relapse after apparently control condition (n ¼ 20), propranolol (40 mg) was administered America,
successful treatment. Once emotional memory is established, it appears to last forever. From an evolutionary perspective, it is extremely Analysis of variance showed fear conditioning on day 1 (stimulus  functional to never forget the most important events in life. However, trial, F1,38 ¼ 46.91, P o 0.001, Z2 ¼ 0.55; ). We observed no the putative indelibility of emotional memory can also be harmful and difference in fear learning between the propranolol and placebo group maladaptive, such as in some trauma victims who suffer from dreadful (stimulus  trial  condition, F1,38 o 1.37; Supplementary Data memories and anxiety. If emotional memory could be weakened or online). On day 2, the two groups expressed comparable levels of startle even erased, then we might be able to eliminate the root of many response during the fear memory reactivation (t38 o 1). In addition, psychiatric disorders, such as post-traumatic stress disorder. Recently, it the conditioned fear memory was equally well consolidated in the two was rediscovered that fear memory in animals is not necessarily groups, as is indicated by both the absence of a main effect of trial from permanent but can change when retriev. The reactivation of a the last three acquisition trials to the reactivation trial (F1,38 o 1) and consolidated (fear) memory can return it to a labile, supposedly protein the absence of a trial  condition interaction effect (F1,38 o 1). These synthesis–dependent state, a process that is referred to as reconsolida- data demonstrate that propranolol did not directly affect the expression tion. Reconsolidation of fear memory can be influenced by neurobio- of the fear memory. Propranolol also did not reduce the startle response logical manipulations during or shortly after the reactivation period per se, as we found no effects of propranolol on the habituation trials These manipulations are thought to alter protein synthesis directlyor (main effect of condition and trial  condition interaction, F1,35 o 1; by interacting with the release of neurotransmitters (for example, norepinephrine) in the amygdala. At the behavioral level, this may In contrast with the pill placebo condition, the administration of lead to changes in later expressions of that fear memory. In particular, propranolol significantly decreased the differential startle response infusion of propranolol into the amygdala of rats shortly after the 48 h later c), that is, from acquisition (trial 6–8, day 1) to reactivation period of a previously acquired fear association impaired extinction (trial 1–3, day 3; stimulus  trial  condition, F1,38 ¼ 17.17, the fear expression on a long-term test. Apparently, propranolol P o 0.001, Z2 = 0.31). Post hoc comparisons showed that propranolol disrupts the reconsolidation of reactivated fear memoriAnimal strongly reduced the expression of fear memory (stimulus  trial, and human studies have shown that adrenal stress hormones activate F1,19 ¼ 25.47, P o 0.001, Z2 ¼ 0.57), whereas the differential startle adrenergic receptors in the amygdala and that the basolateral amygdala response remained stable in the pill placebo condition (stimulus  trial, F1,19 o 1). In the propranolol condition, the conditioned Department of Clinical Psychology, University of Amsterdam, Roetersstraat 15, 1018 WB Amsterdam, The Netherlands. Correspondence should be addressed to M.K.
(.
Received 3 November 2008; accepted 9 January 2009; published online 15 February 2009; VOLUME 12 [ NUMBER 3 [ MARCH 2009 NATURE NEUROSCIENCE All rights reser
Figure 1 Propranolol disrupts the reconsolidation of a fear memory, but not declarative memory. (a–f) Mean startle potentiation to the fear-conditionedstimulus (CS1), the control stimulus (CS2) and noise alone (NA) trials (left) and mean expectancy scores of the unconditioned stimulus to CS1 and CS2 trials(right) during acquisition (trial 1–8), extinction (trial 1–10) and test (trial 1–5) for the placebo (n ¼ 20, a,b), propranolol reactivation (n ¼ 20, c,d) and America,
propranolol without reactivation (n ¼ 20, e,f) group. CS1+ refers to the fear conditioned stimulus during acquisition, CS1– refers to the fear conditionedstimulus during extinction and test, CS1-R refers to the reactivation of the fear conditioned stimulus and CS2– refers to the control stimulus during allphases of the experiment. Error bars represent s.e.m.
fear response was not only reduced but even eliminated, as we (stimulus  trial, F1,18 ¼ 10.33, P o 0.01, Z2 ¼ 0.37; but not in no longer observed the differential startle response (extinction trial the propranolol condition (stimulus  trial, F1,19 o 1; The 1–3, day 3; t19 o 1.22). In contrast, the differential startle response reinstatement procedure did even not reveal any differential startle remained significant in the placebo condition (t19 ¼ 5.26, P o 0.001, response to the first test trial in the propranolol group (t19 o 1).
To determine whether the effect of propranolol requires active Given that the differential startle response was already eliminated in retrieval of the fear memory, we administered propranolol to another the propranolol condition, the two groups differed over the course of fear-conditioned group (n ¼ 20) without memory reactivation. Omis- extinction training on day 3 (stimulus  trial  condition, F1,38 ¼ 5.38, sion of memory reactivation after propranolol intake yielded normal P o 0.05, Z2 ¼ 0.12). Post hoc comparisons showed a significant fear responses and a return of fear 48 h after acquisition (stimulus  decrease of the differential startle response in the placebo condition trial  condition, F1,38 o 1.2; and Supplementary Data).
(stimulus  trial, F1,19 ¼ 11.31, P o 0.005, Z2 ¼ 0.37), but no change of Analysis of variance showed a different pattern for the contingency the differential startle response in the propranolol condition (stimulus learning data, with no effects of propranolol (stimulus  trial   trial, F1,19 o 1) ,c). At the end of extinction (trial 8–10), the differential startle response was still lower in the propranolol condition In sum, oral administration of the b-adrenergic receptor antago- than in the placebo condition (stimulus  condition, F1,38 ¼ 7.94, nist propranolol before reactivation of a fear memory resulted in a substantial weakening of the fear response. We used fear-relevant Exposure to the aversive stimulus (unconditioned stimulus) follow- stimuli (pictures of spiders) because these are especially resistant to ing extinction has been shown to reinstate the expression of the original extinction following fear conditioninEven more notable is our fear memory in animaand humanEvidence for a reinstatement finding that one reactivation trial combined with the administration effect is indicated by an increase of the differential startle response from of propranolol completely eliminated the behavioral expression of the last extinction trials (trial 8–10) to the first test trial. Comparison of the fear memory 24 h later. Second, our finding that a well- the reinstatement effect between the propranolol and placebo condi- established retrieval technique for fear memories (reinstatement) tion showed significantly more reinstatement in the placebo condition failed to uncover any fear response suggests that the fear memory (stimulus  trial  condition, F1,37 ¼ 8.72, P o 0.01, Z2 ¼ 0.19). We may either be erased (storage theory) or may be unavailable as a observed a significant reinstatement effect in the placebo condition result of retrieval failure (retrieval theory)Note that no behavioral NATURE NEUROSCIENCE VOLUME 12 [ NUMBER 3 [ MARCH 2009 procedure is currently available that differentiates between these two understanding and treatment of persistent and self-perpetuating views of amnesia. Notably, the propranolol manipulation left memories in individuals suffering from emotional disorders.
the declarative memory for the acquired contingency between theconditioned and unconditioned stimulus intact, but this knowledge Note: Supplementary information is available on the website.
no longer produced emotional effects. Our finding that propranolol eliminated the fear response, without affecting declarative memory, We thank B. Molenkamp for technical help. This work was supported by a is consistent with the observed double dissociation of fear condition- Vici grant (M.K.) from The Netherlands Organization for Scientific Research.
ing and declarative knowledge relative to the amygdala andhippocampus in humans. Propranolol selectively acts on the AUTHOR CONTRIBUTIONSM.K and M.S. designed the study. M.S. collected data. M.K. and M.S. analyzed b-adrenergic receptors in the amygdala during emotional informa- the data, wrote the initial manuscript and were involved in the revision process.
tion processing in animals and humanIt may be hypothesized All authors discussed the results and commented on the manuscript.
that beta-adrenergic blockade during reconsolidation may selectivelydisrupt the protein synthesis of the amygdalar fear memory, resulting Published online at Reprints and permissions information is available online at in deconsolidation of the fear memory trace while leaving the declarative memory in the hippocampus untouched.
Our findings are consistent with those of a recent preliminary study 1. Bouton, M.E. Biol. Psychiatry 52, 976–986 (2002).
of patients with post-traumatic stress disorder in which post-retrieval 2. Craske, M.G. Anxiety Disorders: Psychological Approaches to Theory and Treatment (Westview Press, Boulder, Colorado, 1999).
propranolol seemed to reduce subsequent physiological responding 3. Dudai, Y. Curr. Opin. Neurobiol. 16, 174–178 (2006).
to traumatic memTogether, these results strongly suggest that 4. Nader, K., Schafe, G.E. & LeDoux, J.E. Nature 406, 722–726 (2000).
b-adrenergic receptors are critically involved in the reconsolidation 5. Tronson, N.C. & Taylor, J.R. Nat. Rev. Neurosci. 8, 262–275 (2007).
6. Canal, C.E. & Gold, P.E. Behav. Neurosci. 121, 732–741 (2007).
process of conditioned fear memories in humans. It is clear that 7. McGaugh, J.L. Annu. Rev. Neurosci. 27, 1–28 (2004).
b-adrenergic blockade during reconsolidation outperformed the tradi- 8. Debiec, J. & LeDoux, J.E. Neuroscience 129, 267–272 (2004).
tional extinction procedure. But most importantly, and in contrast with 9. van Stegeren, A.H. et al. Neuroimage 24, 898–909 (2005).
10. Davis, M. Am. Psychol. 61, 741–756 (2006).
the traditional extinction procedure, disrupting reconsolidation of fear 11. Norrholm, S.D. et al. Learn. Mem. 13, 681–685 (2006).
memory prevented the return of fear. Millions of people suffer from 12. Mineka, S. & Ohman, A. Biol. Psychiatry 52, 927–937 (2002).
13. Nader, K. & Wang, S.-H. Learn. Mem. 13, 530–535 (2006).
emotional disorders and the relapse of fear, even after successful 14. Phelps, E.A. Curr. Opin. Neurobiol. 14, 198–202 (2004).
All rights reser
treatment. Our findings may have important implications for the 15. Brunet, A. et al. J. Psychiatr. Res. 42, 503–506 (2008).
America,
VOLUME 12 [ NUMBER 3 [ MARCH 2009 NATURE NEUROSCIENCE

Source: http://dnitz.com/files/Kindt_extinction_human_propranolol.pdf