What Role Does Serotonin Play in PTSD?

Apr 01, 2006

As many as 90% of Americans are exposed to at least one traumatic event in the course of their lives. Many more are exposed to more than one traumatic event. Short- and long-term sequelae of traumatic exposure vary greatly and range from complete recovery, to severe and debilitating PTSD.

The DSM-IV defines a traumaticevent as one that involves a threatof death or physical integrity toself or others and results in a subjectiveresponse of fear, helplessness orhorror. Epidemiological research disclosedthat up to 90% of US citizensare exposed to at least one traumaticevent in the course of their lives (Breslauand Kessler, 2001), while many moreare exposed to more than one event.

Data gathered from clinical and preclinicalresearch demonstrate a uniquelyhomogeneous response to acute traumaticexposure in both biology andphenomenology. Thus, the initial physiologicalresponse to threat should beconsidered a normal adaptive survivalmechanism. However, short- and long-termsequelae of traumatic exposuregreatly vary, ranging from complete recoveryto severe and debilitating PTSD.

The role of neuroimaging

The increasing availability and advancementof neuroimaging technology providesa solid backbone for ongoingstudies aimed at deciphering neurologicaldifferences between subjectsdiagnosed with psychiatric disorders,such as PTSD, and healthy controls.Existing neuroimaging techniques,including positron emission tomography (PET), single photon computedtomography (SPECT) and functionalmagnetic resonance imaging (fMRI),offer the ability to assess regional cerebralblood flow and glucose metabolismin PTSD subjects in vivo. Previousstudies have included subjects scannedwhile in a resting state; during pharmacologicchallenges; while engagedin cognitive tasks; or experiencing functionalstimuli (such as viewing facesdepicting various emotional states).Functional imaging allows for pairing the visualization of brain activity withparadigms and tasks designed to elicitactivation in specific brain areas, thusparsing out hypothesized differencesbetween subject groups. PET additionallyprovides a modality for examiningneurotransmitter systems. Bystudying specific neuronal pathwaysthroughout the brain that distributeneurotransmitters such as serotonin (5-HT) and dopamine (DA), we are ableto gain an understanding of how alterationsin brain function may contributeto anomalous behavioral traits (eg,PTSD symptomatology).

Imaging studies have already begunhelping researchers distinguish brainpathways and neuronal circuits that maybe associated with behavior, such asthe involvement of the amygdala inemotional processing (Figure 1). Techniquesdiffer in resolution, radioligand availability and analysis methods, providingan array of imaging options.

The latest generation of PET scannersis now capable of distinguishingdifferences in some smaller brainregions such as the raphe and amygdala.Using fMRI, studies in largersamples may provide definitive answersregarding cerebral blood flow differencesbetween PTSD subjects andcontrols. MRI studies may also focuson brain volume comparisons, revealingpossible structural abnormalities (i.e., atrophy) in brain regions associatedwith PTSD.

Ultimately, neuronal circuits are notindependently operating mechanisms,with each translating into one specificbehavioral phenomenon. Rather, theywork in conjunction with multiple pathwaysand neurotransmitter systems.Thus, utilizing various neuroimagingtechniques is useful in identifyingregions of altered activity and maycontribute to understanding the underlyingpathophysiology of PTSD.

Serotonergic mechanismsin PTSD

Different types of acute stress result inincreased 5-HT turnover in the medialprefrontal cortex (mPFC), nucleus accumbens,amygdala and lateral hypothalamusin experimental animals (Inoueet al., 1994). However, exposure to repeated stress within a learned-helplessnessmodel resulted in a decrease of5-HT release in the frontal cortex (possiblyreflecting 5-HT depletion by continuedrelease).

Pretreatment with benzodiazepinereceptor agonists or tricyclic antidepressantdrugs prevents the behavioralsyndrome of learned helplessness andreduces 5-HT release. Treatment withantidepressants such as selective 5-HTreuptake inhibitors, and infusion of 5-HTinto the frontal cortex reverses the behavioralpattern (Petty et al., 1997). Correspondingly,administration of 5-HTreceptor antagonists produces behavioraldeficits resembling those of learnedhelplessness (Petty et al., 1997).

Serotonin release may have bothanxiogenic and anxiolytic effects. Thisapparently depends on the region of theforebrain involved and the receptorsubtype that is predominantly stimulated.Anxiogenic effects are mediatedvia 5-HT2A receptor stimulation;whereas stimulation of 5-HT1A receptorsis anxiolytic and may even provideresilience to aversive events. Postsynaptic5-HT1A receptor gene expressionis under tonic inhibition by adrenalsteroids in the hippocampus, apparentlymostly by mineralocorticoid receptors(Lopez et al., 1998).

The 5HT1A receptor density andmRNA levels decrease in response tostress or cortisol administration and increasefollowing adrenalectomy (Lopezet al., 1998). Stress-induced downregulationof 5-HT1A receptor expressionis prevented by adrenalectomy, showingthe importance of circulating adrenalsteroids in mediating this effect. Thisregulatory steroid effect is rapid, and5-HT1AmRNA levels markedly decreasewithin hours of mineralocorticoid receptorstimulation (Lopez et al., 1998). Conversely,5-HT2A receptor expression isupregulated during chronic stress andcortisol administration and downregulatedin response to adrenalectomy.

There is increasing evidence for abnormalitiesin serotonergic function insubjects with PTSD. Patients withcombat-related PTSD had decreasedplatelet paroxetine (Paxil) binding,suggesting alterations in the 5-HT transporter.Challenge studies probing theserotonergic system using mCPP suggestedthat a subgroup of patients withPTSD develops anxiety and flashbacksupon provocation with this agent(Southwick et al., 1999). However,PTSD symptoms can be elicited alsoby other compounds such as lactate (Rainey et al., 1987). This indicates thatinduction of unspecific anxiety canprovoke symptoms characteristic forPTSD. Davis et al. (1999) used the serotonin-releasing agent and reuptakeinhibitor D-fenfluramine in PTSDpatients and demonstrated a significantlylower prolactin response comparedto control subjects, suggestingcentral serotonergic dysfunction(Southwick et al., 1999).

Several recent studies suggest closeinteractions between serotonergic and γ-aminobutyric acid (GABA)-ergic systems.Mice lacking the 5-HT1A receptordisplay marked anxiety (Heisler et al.,1998; Parks et al., 1998; Ramboz et al.,1998), and animals exposed to stressexhibit downregulation of 5-HT1A receptors(McKittrick et al., 1995). Subordinaterats in a dominance hierarchyshow severe anxiety accompaniedby reduced 5-HT1A receptor levels(McKittrick et al., 1995). More recentlyit was shown that 5-HT1A receptor knockoutmice show: 1) a reduction in the α1and α2 subunits of the GABAA receptorfunction; 2) reduced binding of bothbenzodiazepine and non- benzodiazepineGABAA receptor-ligand; and 3) benzodiazepine-resistant anxiety (Sibille etal., 2000). This suggests a pathologicalpathway originating from 5-HT1A receptordeficit leading towards dysfunctionswithin GABAergic systems, resulting inincreased levels of anxiety.

Therefore, a logical next step in theevaluation of brain systems possibly involvedin the pathophysiology of PTSDwas a study to determine 5-HT1A receptorexpression in patients with PTSDversus controls. We acquired PET images of 5-HT1A receptor binding usingPET imaging on 12 unmedicated PTSDsubjects and 11 healthy controls withouta history of trauma using [18F]fluorocarbonyl-WAY-100635, a highlyselective 5-HT1A receptor radioligand(Bonne et al., 2005).

Unexpectedly, we found no differencein 5-HT1A receptor expression between the groups (Figure 2). Thisresult suggests no direct role for the 5-HT1A receptor in the pathophysiologyof PTSD; however it does not excludeits relevance in mediating the effects ofSSRIs in the treatment of PTSD byinvolving other transmitter systems andneurotrophic systems.

Pharmacologic treatments

It has to be noted that sertraline (Zoloft)is the only FDA-approved medicationfor PTSD. Therefore, all the belowdescribeduses of medications for PTSDare off-label. While treatment with tricyclicantidepressants improved depressive,anxiety and intrusive symptoms,it did not significantly change all the coresymptom clusters in PTSD (Davidsonet al., 1990; Frank et al., 1988; Reist etal., 1989). Two controlled trials withphenelzine (Nardil) showed contradictoryresults (Frank et al., 1988; Shestatzkyet al., 1988), while open trials withpropranolol (Betachron ER, Inderal) andclonidine (Catapres) helped reducesymptoms only in the reexperiencing andhyperarousal clusters.

Although benzodiazepines administeredas single agents have anxiolyticand anti-panic effects, their role in thetreatment of PTSD is limited. Braun et al. (1990) demonstrated a modest, albeitsignificant, improvement in anxiety anddepressive symptoms in 10 Israeli patientswith PTSD using a short-actingbenzodiazepine, alprazolam (Xanax), ina double-blind crossover study. However,there was no significant changein core PTSD symptoms. Administrationof clonazepam (Klonopin) oralprazolam immediately following atraumatic event did not alter the onsetof either anxiety or PTSD symptoms atone and six months (Gelpin et al., 1996).However PTSD patients randomized tobenzodiazepines had a decrease in heartrate over time.

Van der Kolk et al. (1994) performedthe first placebo controlled study of anSSRI (fluoxetine [Prozac]) in PTSD anddemonstrated a significant improvementin all PTSD symptoms. In the largestplacebo-controlled study of sertraline,the drug was shown to be effective intreating avoidance and hyperarousalsymptoms but demonstrated no significanteffect on reexperiencing/intrusioncluster of symptoms (Brady et al., 2000).Marshall and colleagues (2001) randomizedpatients to either 20 mg (n=183) or40 mg (n=182) of paroxetine, or placebo(n=186). Paroxetine-treated patients inboth dosage groups demonstrated significantlygreater improvement on all threeclusters of PTSD (reexperiencing, avoidance/numbing and hyperarousal) comparedto placebo-treated patients at studyend point.

However, one of the main limitationsof SSRI treatment in PTSD is the timelag between initiating treatment andonset of clinical response, which cantake up to six weeks. Furthermore, therecould be an initial exacerbation of anxietysymptoms and insomnia that couldlead to noncompliance or early discontinuationof treatment.


Researchers have successfully determinedthe neurocircuitry underlyingmechanisms, such as fear and anxietyor extinction, which are believed to beabnormal in PTSD. Others have beensuccessful in determining neurochemicalcorrelates of PTSD symptoms.However, we have not yet sufficientlyaddressed the potential role of receptorsand transporters that have been shownto regulate neurochemical mechanismsimportant for PTSD. More recent researchhas focused on the determinationof genes and gene variants which determinetransmitter synthesis and releasein healthy people and individuals sufferingfrom stress-related disorders. Thesestudies will likely lead to an enhancedunderstanding of an individual's vulnerabilityto environmental stressors.

In combination with modern imagingtechniques, including fMRI and molecularimaging using PET, we will havenovel insight into processes underlyingstress-related disorders such as PTSD.Whereas fMRI will provide insight intoneural connectivity of circuits involvedin PTSD, the strength of PET imagingwill be showing the neurochemicalprocesses that underlie altered emotionprocessing or hyperarousal in patientswith PTSD. The ultimate goal of theseresearch efforts is to provide novel,improved treatments for people sufferingfrom trauma-related symptoms.

Dr. Neumeister is associate professor and directorof the Molecular Imaging Program inthe department of psychiatry, Yale UniversitySchool of Medicine. Dr. Neumeister indicatedthat he has nothing to disclose.




Bonne O, Vythilingham M, Bain E et al. (2005), Nochange in serotonin 1A receptor imaging in posttraumaticstress disorder. Am J Psychiatry 162(2):383-385.


Brady K, Pearlstein T, Asnis GM et al. (2000), Efficacyand safety of sertraline treatment of posttraumaticstress disorder: a randomized controlled trial. JAMA283(14):1837-1844 [see comment].


Braun P, Greenberg D, Dasberg H, Lerer B (1990),Core symptoms of posttraumatic stress disorderunimproved by alprazolam treatment. J ClinPsychiatry 51(6):236-238.


Breslau N, Kessler RC (2001), The stressor criterionin DSM-IV posttraumatic stress disorder: an empiricalinvestigation. Biol Psychiatry 50(9):699-704.


Davidson J, Kudler H, Smith R et al. (1990), Treatmentof posttraumatic stress disorder with amitriptyline and placebo. Arch Gen Psychiatry 47(3):259-266.


Davis LL, Clark DM, Kramer GL et al. (1999), Dfenfluraminechallenge in posttraumatic stress disorder.Biol Psychiatry 45(7):928-930.


Frank JB, Kosten TR, Giller EL Jr, Dan E (1988), Arandomized clinical trial of phenelzine and imipraminefor posttraumatic stress disorder. Am JPsychiatry 145(10):1289-1291.


Gelpin E, Bonne O, Peri T et al. (1996), Treatmentof recent trauma survivors with benzodiazepines: aprospective study. J Clin Psychiatry 57(9):390-394.


Heisler LK, Chu HM, Brennan TJ et al. (1998),Elevated anxiety and antidepressant-like responsesin serotonin 5-HT1A receptor mutant mice. Proc NatAcad Sci U S A 95(25):15049-15054 [see comment].


Inoue T, Tsuchiya K, Koyama T (1994), Regional changesin dopamine and serotonin activation with various intensityof physical and psychological stress in the rat brain.Pharmacol Biochem Behav 49(4):911-920.


Lopez JF, Chalmers DT, Little KY, Watson SJ (1998),A.E. Bennett Research Award. Regulation of serotonin1A,glucocorticoid, and mineralocorticoid receptor in rat andhuman hippocampus: implications for the neurobiologyof depression. Biol Psychiatry 43(8):547-573.


Marshall RD, Beebe KL, Oldham M, Zaninelli R(2001), Efficacy and safety of paroxetine treatmentfor chronic PTSD: a fixed-dose, placebo-controlledstudy. Am J Psychiatry 158(12):1982-1988.


McKittrick CR, Blanchard DC, Blanchard RJ et al. (1995),Serotonin receptor binding in a colony model of chronicsocial stress. Biol Psychiatry 37(6):383-393.


Parks CL, Robinson PS, Sibille E et al. (1998), Increasedanxiety of mice lacking the serotonin1A receptor. ProcNat Acad Sci U S A 95(18):10734-10739.


Petty F, Kramer GL, Wu J (1997), Serotonergicmodulation of learned helplessness. Ann N Y AcadSci 821:538-541.


Rainey JM Jr, Aleem A, Ortiz A et al. (1987), A laboratoryprocedure for the induction of flashbacks. AmJ Psychiatry 144(10):1317-1319.


Ramboz S, Oosting R, Amara DA et al. (1998),Serotonin receptor 1A knockout: an animal modelof anxiety-related disorder. Proc Natl Acad Sci U SA 95(24):14476-14781 [see comment].


Reist C, Kauffmann CD, Haier RJ et al (1989), Acontrolled trial of desipramine in 18 men with posttraumaticstress disorder. Am J Psychiatry 146(4):513-516 [see comments].


Shestatzky M, Greenberg D, Lerer B (1988), Acontrolled trial of phenelzine in posttraumatic stressdisorder. Psychiatry Res 24(2):149-155.


Sibille E, Pavlides C, Benke D, Toth M (2000), Geneticinactivation of the serotonin1A receptor in miceresults in downregulation of major GABAA receptorsubunits, reduction of GABAA receptor binding, andbenzodiazepine-resistant anxiety. J Neuroscience20(8):2758-2765.


Southwick SM, Paige S, Morgan CA 3rd et al. (1999),Neurotransmitter alterations in PTSD: catecholaminesand serotonin. Semin Clin Neuropsychiatry4(4):242-248.


van der Kolk BA, Dreyfuss D, Michaels M et al (1994),Fluoxetine in posttraumatic stress disorder. J ClinPsychiatry 55(12):517-522 [see comments].