Diversos estudos demonstram que o estresse infantil, incluindo situações de perda dos pais, negligência e abusos, representa um forte fator de risco para o desenvolvimento de transtornos de ansiedade, sendo de especial interesse para este trabalho, o transtorno do pânico. Modelos de estresse neonatal em animais de laboratório, que se baseiam na ruptura da relação mãe-filhote, como a separação materna e a privação materna, têm sido amplamente utilizados para avaliar as consequências desse estressor sobre a expressão de comportamentos defensivos associados à ansiedade na vida adulta. No entanto, pouco se sabe sobre seus efeitos em modelos animais de ataques de pânico, mais especificamente aqueles que associam esta condição emocional à resposta defensiva de fuga em animais. Diante disso, o objetivo inicial do presente trabalho foi o de estender as investigações dos efeitos do estresse neonatal sobre o comportamento de fuga de ratos adultos (após 60 dias de nascimento) observado no labirinto em T elevado (LTE), pela estimulação elétrica da substância cinzenta periaquedutal (SCPD) e durante a exposição a um ambiente em hipóxia (7% O2). Para efeitos comparativos, esses animais também foram testados em modelos animais associados à ansiedade generalizada e a depressão. Observamos que ratos Wistar submetidos à separação materna (3h/dia, do 2º ao 21º dia pós-nascimento) não diferiram de animais controles nos parâmetros comportamentais analisados nos modelos de pânico (fuga no LTE e pela estimulação elétrica da SCPD), nos de ansiedade (resposta de esquiva no LTE e o beber punido no teste de conflito de Vogel) ou no de depressão (tempo gasto em imobilidade no teste do nado forçado). Já em ratos privados da mãe (por 24h no 11º dia pós- nascimento), embora este estressor não tenha alterada a resposta de fuga no LTE, ele aumentou a expressão deste comportamento durante a exposição à hipóxia, sugestivo de um efeito panicogênico. Ainda empregando a privação materna, observamos que a administração intraperitoneal de um inibidor da síntese de serotonina, a pclorofenilalanina metil éster (p-CPA - 100mg/Kg/dia, por 4 dias antes dos testes comportamentais) facilitou a expressão do comportamento de fuga durante o teste da hipóxia nos animais controle, de maneira semelhante ao efeito obtido somente com a privação materna. Porém este tratamento não potencializou a fuga promovida pela privação materna. Já os níveis plasmáticos de corticosterona foram aumentados pela exposição à hipóxia, independentemente dos animais terem sido previamente privados da mãe ou terem recebido o pCPA antes do teste. Por fim, também observamos, através de uma análise por Western Blotting, que nem a privação materna ou a exposição à hipóxia altera a concentração de receptores serotonérgicos do tipo 5-HT1A na SCPD ou na amígdala. Em suma, nossos resultados mostram que a privação materna promove uma facilitação da resposta de fuga na hipóxia, sugerindo uma relação entre esse estresse neonatal e o desencadeamento de ataques de pânico de um subtipo específico, o pânico respiratório. Contudo, no que diz respeito ao envolvimento da neurotransmissão serotonérgica, mais estudos são necessários para entender sua participação nessa resposta. / Early life stress (ELS), including parental loss due to death, neglect or abuse, represents a major risk factor for the late development of psychiatric disorders, such as anxiety disorders. Animal models of ELS that are based on the disruption of mother-infant relationship, such as the repeated maternal separation or maternal deprivation, have been extensively used for the investigation of the longterm effects of these stressors on the expression of defensive behaviors associated with anxiety. However, little is known about their effects on animal models of panic attacks, more specifically in those that associate this emotional condition with escape behavior in animals. Therefore, the aim of the present study was to extend the investigation on the long-term consequences of neonatal stress on the escape response of adult rats (60 days after birth) evoked by the elevated T maze (ETM), electrical stimulation of the dorsal periaqueductal gray (DPAG) or hypoxia (7% O2). For comparative reasons, these animals were also tested in animal models of anxiety and depression. The results showed that the repeated maternal separation of male Wistar (3 hours/day from day 2 to 21 after birth) did not affect the behavioral indexes measured in the panic (escape in the ETM or after DPAG electrical stimulation), anxiety (ETM inhibitory avoidance or punished licking in the Vogel conflict test) or depression (time in immobility in the forced swimming test) models. On the other hand, rats submitted to maternal deprivation (24 hs in the 11th day after birth), although not differing from the control animals on escape expression in the ETM, showed a pronounced escape response during hypoxia, indicating a panicogenic-like response. Also, using this maternal deprivation protocol, we observed that systemic administration of a 5-HT synthesis inhibitor, p-chlrophenilalanine metylester (p-CPA - 100mg/Kg/day, for 4 days before the behavioral tests), facilitated escape expression during hypoxia in non-deprived animals to a level observed in non-pharmacologically treated deprived animals. We also observed that plasma corticosterone levels were increased 30 minutes after hypoxia exposure, independently of the previous condition of the animals (deprivation or drug treatment). Finally, we observed that the number of 5-HT1A receptors in the DPAG or amygdala, measured by Western Blotting, was not affect by previous maternal deprivation or exposure to hypoxia. Taken together, our results show that a single maternal deprivation episode facilitates the expression of escape behavior during hypoxia, suggesting a relationship between this ELS with the observation of a specific subtype of panic attack, the respiratory panic. Further studies are required in order to clarify the 5- HT involvement on these responses.

Abstract Early risk factors of the antenatal period and infancy have been increasingly linked to psychiatric disorders. The aim of this thesis was to study the associations between very early parental separation and maternal depressed mood in pregnancy on the other hand, and schizophrenia and criminality in the offspring in adolescence and adulthood, on the other, in two data sets. In the Christmas Seal Home Children Study the index cohort consisted of 3 020 subjects born in Finland in 1945–65 who were temporarily isolated from their family immediately after birth to nursing homes, the Christmas Seal Homes, due to tuberculosis in the family. The average separation time was seven months. For every index subject, two reference subjects were matched for sex, year of birth and place of birth. Data were obtained on schizophrenia from the Finnish Hospital Discharge Register (FHDR) in 1971–98 and on criminal offences from Statistics Finland in 1977–98. The 28-year cumulative incidence of schizophrenia was 1.6% both in the index cohort and in the reference cohort (RR 1.0; 95% CI 0.8–1.4). Both male and female index subjects had committed crimes more commonly than the reference subjects (in men RR 1.3; 95% CI 1.2–1.4; in women RR 1.5; 1.2–2.0). Of the male index subjects 12.1% as compared with only 7.1% of the reference cohort had committed violent offences (RR 1.7; 1.4–2.1). In the Northern Finland 1966 Birth Cohort mothers of 12 058 babies were asked at mid-gestation at the antenatal clinic if they felt depressed. This general population birth cohort of the children was followed up for 31 years being record-linked with the FHDR covering the years 1982–97 and with the criminal register of the Ministry of Justice up to 1998. We divided the schizophrenia patients into those having a psychotic first-degree relative (schizophrenia patients with familial risk for psychosis FR) and those without one. The cumulative incidence of hospital-treated schizophrenia was 1.3% among the offspring of depressed mothers and 0.9% among the descendants of non-depressed mothers (RR 1.5; 95% CI 0.9–2.4). The prevalence of antenatal depression was 35% in mothers of schizophrenia patients with FR. The respective prevalence was 14% both in the mothers of schizophrenia patients without FR and in the mothers of other cohort members. Both male and female offspring of antenatally depressed mothers were more commonly criminal offenders than offspring of non-depressed mothers (in men adjusted OR 1.5; 95% CI 1.2–1.9; in women OR 1.5; 0.8–3.0). In males, 6.5% with depressed mothers and 3.2% with non-depressed mothers had committed violent offences (adjusted OR 1.6; 1.1–2.4). Very early separation and mothers' depressed mood in pregnancy are per se unlikely to increase the risk for schizophrenia in the offspring, but seem to be connected to criminal behaviour, especially violent criminality in men. / Tiivistelmä Raskaus- ja imeväisajan varhaiset tekijät on lisääntyvästi yhdistetty lapsen tuleviin mielenterveyshäiriöihin. Tarkoituksena oli tutkia hyvin varhaisen eron (separaation) ja äidin raskaudenaikaisen masentuneen mielialan yhteyttä lasten skitsofreniaan ja rikollisuuteen nuoruudessa ja aikuisuudessa kahdessa eri aineistossa. Joulumerkkikoti-lasten tutkimuksessa indeksikohortti koostui 1945–65 syntyneistä 3 020 tutkittavasta, jotka erotettiin väliaikaisesti perheistään heti syntymän jälkeen hoitokoteihin, Joulumerkkikoteihin, perheen tuberkuloosin takia. Ero vanhemmista kesti keskimäärin seitsemän kuukautta. Jokaiselle indeksitutkittavalle valittiin kaksi sukupuolen, syntymävuoden ja -paikan mukaan kaltaistettua verrokkitutkittavaa. Tieto skitsofreniaan sairastumisesta hankittiin sairaaloiden poistoilmoitusrekisteristä vv. 1971–98 ja rikoksista Tilastokeskuksesta 1977–98. 28 vuoden kumulatiivinen sairastuvuus skitsofreniaan oli 1,6 % sekä indeksi- että verrokkikohortilla (riskisuhde RR 1.0; 95 %:n luottamusväli CI 0,8–1,4). Sekä miehistä että naisista indeksitutkittavat olivat tehneet useammin rikoksia kuin vertailuryhmä (miehillä RR 1,3; 1,2–1,4; naisilla RR 1,5; 1,2–2,0). Miehistä 12,1 % indeksitutkittavista ja vain 7,1 % vertailuryhmästä oli tehnyt väkivaltarikoksen (RR 1,7; 1,4–2,1). Pohjois-Suomen 1966 syntymäkohortin 12 058 lapsen äideiltä kysyttiin keskiraskauden aikana äitiysneuvolassa, kokivatko he mielialansa masentuneeksi. Tämän väestötason syntymäkohortin (siis lasten) tietoja hankittiin 31-vuotisseurannassa sairaaloiden poistoilmoitusrekisteristä vuosilta 1982–97 ja oikeusministeriön rikosrekisteristä vuoteen 1998. Skitsofreniaan sairastuneet jaettiin niihin, joiden 1. asteen sukulainen oli ollut / ei ollut ollut psykoottinen. Sairaalahoitoa vaatineen skitsofrenian kumulatiivinen sairastuvuus oli 1,3 % masentuneiden ja 0,9 % masentumattomien äitien lapsilla (RR 1,5; 0,9–2,4). Raskaudenaikaisen masentuneen mielialan esiintyvyys oli 35 % niiden skitsofreniapotilaiden äideillä, joilla oli ollut lähisuvussa psykoosia. Vastaavasti masentunutta mielialaa esiintyi 14 %:lla sekä niiden skitsofrenia-potilaiden äideistä, joilla ei ollut sukurasitusta, että muiden kohorttitutkittavien äideistä. Sekä masentuneiden äitien pojista että tyttäristä useampi oli tehnyt rikoksen kuin masentumattomien äitien lapset (miehillä vakioitu vedonlyöntisuhde OR 1,5; 1,2–1,9; naisilla OR 1,5; 0,8–3,0). Masentuneiden äitien pojista 6,5 % ja masentumattomien äitien pojista 3,2 % oli tehnyt väkivaltarikoksen (vakioitu OR 1,6; 1,1–2,4). Hyvin varhainen ero ja äidin masentunut mieliala raskauden aikana eivät todennäköisesti sinänsä lisää skitsofrenian vaaraa lapsilla, mutta näyttävät olevan yhteydessä lasten rikolliseen käyttäytymiseen, erityisesti väkivaltarikoksiin miehillä.