探討青少年期捆綁處理對海馬迴功能之長期不良影響

全文

(1)國⽴臺灣師範⼤學理學院⽣命科學系碩⼠論⽂ Department of Life Science College of Science. National Taiwan Normal University Master’s Thesis. 探討青少年期捆綁處理對海⾺迴功能之⾧期不良影響 Juvenile immobilization treatment elicits long-term adverse effects on the hippocampal function. 董佑萱 Yu-Hsuen Tung 指導教授﹕呂國棟博⼠ Advisor: Kwok-Tung Lu, Ph.D.. 中華民國 109 年 8 ⽉ August 2020.

(2) ACKNOWLEDGEMENT 時光⾶逝、歲⽉如梭，轉眼間碩班三年的研究所⽣涯即將邁⼊尾聲，驀然回⾸⼼中充滿不捨，感謝讓我有所蛻變的⼀切。在學習的過程中，受到許多⼈的⽀持以及⿎勵，讓我在三年的研究⽣涯充滿了歡笑與⾟勞，也學習到了許多知識以及研究精神。研究論⽂能得以順利完成，⾸先必須感謝我的指導教授呂國棟⽼師，當我因受傷在家休養時，會抽空並利⽤線上上課的⽅式幫我補課。感謝⽼師在論⽂撰寫期間給予的教導與督促，並且在忙碌的教學⼯作中擠出時間來審查、修改我的論⽂。感謝我的⼝試委員嘉義⼤學楊奕玲⽼師和吳游源⽼師以及中正⼤學陳永恩⽼師，百忙之中特地來到台北擔任⼝試委員，還不吝花時間修改我的論⽂，並從中給予許多建議，使得這篇論⽂能更加完善。感謝三年中陪伴在我⾝邊的同學、朋友，感謝他們為我提出的有益的建議和意⾒，有了他們的⽀持、⿎勵和幫助，我才能充實的度過了三年的學習⽣活。特別感謝維星學⾧，我每次遇到難題，總會抽空與我討論，然後⼀起商量解決的辦法，還教我⽤嚴謹的研究態度與精益求精的研究精神去完成實驗。感謝⼦雅學姊，從碩⼀下成為電⽣理. i.

(3) 實驗的夥伴之後彼此合作無間，完成許多困難。學姊畢業後還不忘疼我，經常做便當，讓我每天都有豐盛的午餐可以享⽤。感謝在整個學習期間與我密切合作的鵬愷和給予過我幫助的夥伴們，在此，再⼀次真誠地向幫助過我的⽼師和同學表⽰感謝。因為有各位的關懷與⿎勵，使我在學習的路程中⼜向前邁進的⼀步。. 研究⽣. 董佑萱. 謹誌. 國⽴臺灣師範⼤學⽣命科學系中華民國 109 年 8 ⽉. ii.

(4) ABSTRACT IN CHINESE 青春期遭遇性騷擾(sexual harassment)、學校霸凌(bullying)、情感虐待(psychological abuse)和家庭暴⼒(domestic violence)等創傷經驗會導致⼼理創傷(psychological trauma)，已被證實會對⼼理健康產⽣⾧期的不良影響，並增加成年後罹患精神疾病的⾵險。不幸的是，針對其不良影響的發⽣，⽬前僅有少數研究與有限的治療策略。本研究之主要⽬的為利⽤青春期綑綁處理動物模式(juvenile immobilization treatment, J_IMO)模擬青少年期創傷經驗，以探究其對海⾺迥功能 (hippocampal function)的不良影響及其神經機。本研究採⽤ C57BL / 6J 品系雄性⼩⿏，在出⽣後第 35 天和第 36 天(postnatal day, PND)時進⾏每天⼀次 2 ⼩時，共兩天的 J_IMO 處理。於第⼆次綑綁處理後⼀週時(PND-42)分別進⾏⾏為(behavioral)、電⽣理(electrophysiological)和⽣化(biochemical)實驗。⾏為實驗採⽤抑制迴避測試(inhibitory avoidance, IA) 來評估海⾺迴(hippocampus)的學習記憶功能，IA 為⼀與海⾺迥功能⾼度相關的記憶試驗。我們也利⽤開放空間測試(open field test, OFT)和⾼架⼗字迷宮測試(elevated plus-maze test , EPM) 來評估 J_IMO 的類焦慮⾏為 (anxiety-like behavior)。部分動物則斷頭犧牲取腦，製成海⾺迴腦切⽚⽤於離體胞 iii.

(5) 外電⽣理記錄(in vitro extracellular recording)，以⾼頻電刺激(highfrequency stimulation, HFS)誘發的⾧期增益效應(long-term potentiation, LTP)，來評估動物海⾺迴的神經可塑性(neuroplasticity)。為了防⽌⾏為實驗對⽣化分析所造成之⼲擾，我們以相同的⽅式準備了另外⼀批動物，於 PND-42 時斷頭犧牲，取出其海⾺迴進⾏即時定量聚合酶連鎖反應(real-time polymerase chain reaction, qPCR)分析 Grin2a （NR2A 受體），Slc12a2 （鈉鉀氯共轉運蛋⽩ 2，⼜簡稱 NKCC1），Slc12a5 (鈉鉀氯共轉運蛋⽩ 5，⼜簡稱 KCC2), Gabra1（GABAA 受體）和 Fkbp5 （co-chaperone with the heat shock protein 90 and steroid complex）等基因的表現，並且利⽤西⽅墨點法（Western blot）來確認 NKCC1 的蛋⽩質表現量。實驗結果顯⽰，經 J_IMO 處理的⼩⿏，其抑制性迴避學習增強，合併出現削減學習(extinction learning)減弱的現象。⽽ OFT 的結果顯⽰ J_IMO 組的類焦慮⾏為有增加的情形，然⽽在 EPM 中並未獲得⼀致的結果。J_IMO 組動物的海⾺迴 HFS-LTP 有增強的現象，與對照組相較經 J_IMO 處理⼩⿏的輸⼊和輸出曲線⽐值（input/output curve ratio, I/O curve）中出現顯著增加的現象。⽽在配對脈衝促進（pair-pulse facilitation. PPF）的結果並無顯著差異。這些結果暗⽰海⾺迴 HFS-LTP iv.

(6) 的增強應是源⾃突觸後的機制(post-synaptic mechanism)，例如受體表現量增加(up-regulation)，或者信號傳遞活性增強。qPCR 的結果顯⽰，經 J_IMO 處理⼩⿏海⾺迴中 Grin2a 和 Slc12a2 的表現量顯著增加，可以證明 J_IMO 組海⾺迴突觸後的機制改變。Fkbp5 ， Slc12a5 和 Gabra1 的表現量未呈顯著差異。之後，我們以表⾯灌流(suprafusion) ⽅式投予兩種環形利尿劑(loop diuretics)，呋塞⽶(furosemide)或布美他尼(bumetanide)為 NKCC1 抑制劑，可將海⾺迴 HSF-LTP 恢復⾄正常範圍。綜合各項實驗結果，經過 J_IMO 處理的⼩⿏表現出異常的⾏為表現，包括迴避學習之增強、消減學習(extinction learning)能⼒減弱，和類焦慮⾏為的增加。J_IMO 組的海⾺迴 HSF-LTP 增強海，⾺迴功能相關基因如 Grin2a，Slc12a2 的表現量明顯改變。這些結果共同顯⽰出，給予 J_IMO 急性處理也可能對海⾺迴功能產⽣⾧期影響。本研究結論為 J_IMO 處理成功地模擬了青少年的創傷經歷，並且產⽣⾧期的不良影響，這和前⼈之成年 IMO 處理的研究發現是⼀致的。有趣的是，我們發現 NKCC1 在 J_IMO ⼩⿏中的海⾺迴表現量發⽣了變化，這解釋 J_IMO 治療處理不良反應的可能機制。我們建議 NKCC1 抑制劑如布美他尼(bumetanide)可以作為治療藥物，以減輕青少年創傷事件引起的⾏為異常。 v.

(7) 關鍵字：青春期、綑綁處理、抑制迴避試驗、開放空間試驗、⾼架⼗字迷宮、消減學習、類焦慮⾏為、海⾺迴、胞外記錄法、⾧期增益效益、鈉鉀氯共轉運蛋⽩、環形利尿劑、布美他尼、呋塞⽶。. vi.

(8) ABSTRACT IN ENGLISH Traumatic events during adolescence such as sexual molest, school bullying, emotional abuse, and domestic violence might result in psychological trauma and a long-term deleterious effect on mental health. They would eventually increase the risk of having psychiatry diseases in the adulthood. Unfortunately, the detailed related neural mechanism remains unclear, and limited therapeutic strategies are available to prevent the adverse consequences. Here, a modified juvenile immobilization treatment (J_IMO) animal model was applied to investigate the neural mechanism underlying the adverse effect of juvenile traumatic events on the function of the hippocampus. Briefly, C57BL/6J mice received J_IMO treatment at the postnatal day 35 and 36 (PND). One week later (PND-42), they were subjected to behavioral, electrophysiological, and biochemical experiments. We evaluated the hippocampal function by using an inhibitory avoidance test (IA), a well-known hippocampus-dependent memory task. The anxiety-like behaviors were examined by the open field test (OFT) and elevated plus-maze test (EPM). We also used the in vitro extracellular recording to study the high-frequency stimulation-induced long-term potentiation (HFS-LTP) of the hippocampus. Furthermore, the expression of anxiety-related genes such as Grin2a (NR2A receptor), Gabra1 (GABAA receptor), Slc12a2 (sodium potassium chloride cotransporter-1, NKCC1), Slc12a5 (potassium-chloride transporter member 5, KCC2), and Fkbp5 (co-chaperone with the heat shock protein 90 and steroid complex, FKBP5) of the six weeks old J_IMO mice was vii.

(9) determined by using real-time polymerase chain reaction (qPCR) and western blot. Results showed that enhanced avoidance learning in the J_IMO treated mice. Also, the J_IMO-treated male mice displayed some degree of interference on the extinction of the IA task. Furthermore, an elevation of anxiety-like behavior was revealed in the test of OFT. The hippocampal HFS-LTP increased in the J_IMO treated group, which could explain the enhanced avoidance learning. A significant difference was found only in the input-output curves (I/O curve), but not in the pair-pulse facilitation (PPF), between control group and J_IMO treated mice. These results implied the enhanced hippocampal HFS-LTP was resulted from a postsynaptic mechanism such as increased receptor expression, or an increase of signal transduction activity. Our qPCR results showed that the expression of Grin2a and Slc12a2 were increased significantly, On the contrary, the expression of Fkbp5, Gabra1 and Slc12a5 were not significantly altered in the J_IMO treated mice. Administration of loop diuretics furosemide or bumetanide, NKCC1 inhibitors, restored the hippocampal HSF-LTP to the normal range. Based on our findings, the J_IMO treated mice actually revealed certain abnormal behavioral phenotypes, including a deficit in the extinction of avoidance learning, an increase in anxiety-like behavior, the enhancement of hippocampal HSF-LTP, and a significant upregulation of anxiety-related genes, Grin2a and Slc12a in hippocampus. These results. viii.

(10) collectively indicated that even an acute J_IMO treatment might elicit a long-term impact on hippocampal functions. Conclusively, the J_IMO treatment designed in this study has successfully simulated the juvenile traumatic experience, and has provoked a long-term deleterious impact consistent to the previous finding in adult IMO model studies. It is worth noting that the hippocampal expression of NKCC1 was altered in the J_IMO mice, and that might account for the long-term adverse effect of J_IMO treatment. Accordingly, the NKCC1 inhibitor bumetanide might become a therapeutic agent to relieve the behavioral abnormalities induced by juvenile traumatic events.. Keywords: Juvenile, immobilization treatment, inhibitory avoidance, open field testing, elevated plus maze, extinction, anxiety-like behavior, hippocampus, extracellular recording, long-term potentiation, cationchloride cotransporters, furosemide, bumetanide.. ix.

(11) Table of Contents ACKNOWLEDGEMENT ....................................................................... i ABSTRACT IN CHINESE .................................................................... iii ABSTRACT IN ENGLISH ................................................................... vii ABBREVIATION .................................................................................. xv INTRODUCTION ................................................................................... 1 Adverse juvenile experiences and the associated long-term impact ..... 1 Anxiety disorder and the dysfunction of the HPA axis.......................... 2 Brain structures correlated to the stress-induced behavioral abnormalities ......................................................................................... 5 Glucocorticoid effects on the function of the hippocampus .................. 6 The related genes for stress-induced behavioral abnormalities related genes ...................................................................................................... 8 The impact of stress on the excitatory neurotransmission ................... 10 Stress-induced abnormality of cation-coupled chloride transporters (CCCs) ................................................................................................. 11 Research aim and significance ............................................................ 14 MATERIALS AND METHODS .......................................................... 15 Animals ................................................................................................ 15. x.

(12) Juvenile immobilization treatment (J_IMO) ....................................... 15 Inhibitory avoidance test (IA).............................................................. 16 Open field test (OFT) .......................................................................... 17 Elevated plus-maze (EPM) .................................................................. 18 Brain slice extracellular recording ....................................................... 19 Administration of bumetanide and furosemide ................................... 20 Real-time polymerase chain reaction .................................................. 21 Western blot ......................................................................................... 24 Statistics ............................................................................................... 25 RESULTS ............................................................................................... 27 Experiment-1: Examine the juvenile immobilization treatment effects on the inhibitory avoidance learning. .................................................. 27 Experiment-2: Evaluation of the juvenile immobilization treatment induced anxiety-like behavior in the male mice by open field test and elevated plus-maze test. ....................................................................... 30 Experiment-3: Determine the hippocampal long-term potentiation of the IMO animals using brain slice extracellular recording.................. 32 Experiment-4: Determine the input-output curve and paired-pulse facilitation of the brain slices of J_IMO treated animals..................... 34 Experiment-5: Examine the expression of HPA axis-related and neurotransmitter receptor genes of the hippocampal on the juvenile xi.

(13) IMO animals by using real-time polymerase chain reaction. .............. 36 Experiment-6: Analyze the hippocampal NKCC1 expression of J_IMO mice by Western blotting. .................................................................... 38 Experiment-7: The enhanced hippocampal HFS-LTP of the J_IMO animals were reversed by suprafusion of NKCC1 antagonist in a dosedependent manner. ............................................................................... 39 DISCUSSION ......................................................................................... 43 Summary of the results ........................................................................ 43 The juvenile IMO treatment reveals an opposite effect on the acquisition and the extinction of inhibitory avoidance memory.......... 45 The high-frequency stimulation-induced hippocampal long-term potentiation is enhanced in juvenile IMO treated mice. ...................... 51 The involvement of NKCC1 in the J_IMO treatment effects on hippocampal function. ......................................................................... 54 Conclusion ........................................................................................... 61 REFERENCES ...................................................................................... 63 FIGURES ............................................................................................... 87 Figure 1. Neural circuits in the brain for fear regulation and posttraumatic stress disorder. ..................................................................... 88 Figure 2. The molecular mechanism of BDNF related stress reactions induced by glucocorticoids. ................................................................. 90 xii.

(14) Figure 3. The molecular mechanism of FKBP5 related stress reactions induced by glucocorticoids. ................................................................. 92 Figure 4. Developmental alteration in cation-chloride cotransporter expression during neuronal maturation. .............................................. 94 Figure 5. The behavioral instrument of in Inhibitory avoidance test .. 95 Figure 6. The behavioral instrument of open field test ........................ 96 Figure 7. The behavioral instrument of elevated plus-maze................ 97 Figure 8. Enhanced avoidance learning in the J_IMO treated mice. ... 99 Figure 9. Juvenile immobilization treatment male mice induced anxiety-like behavior evaluated by open field test and elevated plus maze. .................................................................................................. 102 Figure 10. High-frequency stimulation-induced hippocampal LTP of the control and J_IMO treated mice. ................................................. 104 Figure 11. Comparison of the input-output curves and paired-pulse facilitation between the control and juvenile IMO treated mice. ...... 106 Figure 12. Real-time polymerase chain reaction analysis of the expression of HPA axis-related, neurotransmitter, and cation-chloride cotransporters genes alteration. ......................................................... 108 Figure 13: The cation-coupled chloride transporters protein NKCC1 expression was increased in the hippocampus of J_IMO-treated mice. ........................................................................................................... 110 Figure 14. Suprafusion of bumetanide and furosemide blocked the xiii.

(15) formation of LTP in the hippocampus. .............................................. 114. xiv.

(16) ABBREVIATION J_IMO:. Juvenile immobilization treatment. PND:. Postnatal day. PTSD:. Post-traumatic stress disorder. DSM-5:. Diagnostic and Statistical Manual of Mental Disorders. PVN:. Paraventricular nucleus. CRH:. Corticotropin-releasing hormone. ACTH:. Adrenocorticotropic hormone. CNS:. Central nervous system. BDNF:. Brain-derived neurotrophic factor. IA:. Inhibitory avoidance. OFT:. Open field test. EPM:. Elevated plus-maze test. HFS:. High-frequency stimulation. LTP:. Long-term potentiation. fEPSP:. Field excitatory postsynaptic potential. I/O curve:. Input/output curve ratio xv.

(17) PPF:. Pair-pulse facilitation. qPCR:. Real-time polymerase chain reaction. Gapdh:. Glyceraldehyde 3-phosphate dehydrogenase. FkbpP5:. FK506 binding protein 5. Gabra1:. Gamma-aminobutyric acid (GABA) A receptor, subunit alpha 1. Grin2a:. Glutamate receptor, ionotropic, N-methyl-D-aspartate epsilon 1 (NMDA2A). Slc12a2:. Na-K-Cl cotransporter 1 (solute carrier family 12, member 2). Slc12a5:. K-Cl transporter member 5 (solute carrier family 12, member 5). GCs:. Glucocorticoids. NE:. Norepinephrine. FC:. Fear conditioning. Bumetanide: Na+/2Cl−/K+ (NKCC) symporter inhibitor Furosemide: Na+/2Cl−/K+ (NKCC) symporter inhibitor; also, GABAA antagonist. xvi.

(18) INTRODUCTION Adverse juvenile experiences and the associated long-term impact Child abuse is a common social issue and has become a severe burden of social welfare. In USA alone, approximately 700,000 children per year are recognized by child protection services as the victims of abuse (US Department of Health and Human Services, 2017). In 2019, the Department of Statistics to Ministry of Health and Welfare of Taiwan against that children or adolescents suffer from physical, mental, sexual, or negligent treatment. The number of the abused child and adolescent demographic data shows that the number of people from 9 years old to under 18 years old is double that from 0 years old to under nine years old. The proportion of physical abuse is about triple of that of mental abuse (Taiwan health and welfare report- Ministry of Health and Welfare, 2019). Adolescence represents the period of time children are developing into abuts (Arnett, 2000; Casey et al., 2002; Hueston et al., 2017). It is considered a critical period for the programming of future adult behaviors (Sawyer et al., 2012; Hueston et al., 2017). Early life stress was known to be one of the leading risk factor for developing mental problems (Nishi et al., 2014). It could give rise to a significant impact on the mental health, stress response, and physical health in adulthood (Caspi et al., 2003; Nelson et al., 2007; Miller et al., 2011). Many studies have also found that adverse childhood experiences might increase the risk of having learning and memory deficits (Reincke & Hanganu-Opatz, 2017), eating disorders 1.

(19) (Jahng, 2011), metabolic abnormalities (Plotsky & Meaney, 1993), inflammation and high blood pressure (Danese et al., 2009; Stein et al., 2010). Furthermore, increases of the susceptibility in having mood and anxiety disorders (Penza et al., 2003). Clearly, it is an urgent demand in realizing the mechanism underlying the long-term adverse effect of early life stress. In the past three decades, accumulated pieces of evidence have suggested that dysfunction of the hypothalamic-pituitary-adrenal axis (HPA axis), an essential neurohormone circuitry, might also be responsible for the long-term deleterious impact of juvenile adverse experience.. Anxiety disorder and the dysfunction of the HPA axis Anxiety disorder has the highest incidence rate in all the mental illness globally (Strohle et al., 2018). Its symptoms include fear, tachycardia, tremors, hyperventilation, and other physiological responses. There are several different forms of anxiety disorders, such as generalized anxiety disorder, panic disorder, social phobia, obsessive-compulsive disorder (OCD) (Stein et al., 1996; Safren et al., 2002), and post-traumatic stress disorder (PTSD) (Heim & Nemeroff, 2001). The American Psychiatric Association revised the diagnostic criteria of PTSD in the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) for disorders as trauma or stressor-related trauma (American Psychiatric Association, 2013). According to the description of DSM-5, PTSD has eight diagnostic criteria including (1) The source of stress must be directly or indirectly exposed to 2.

(20) the traumatic event and cause syndromes and symptoms; (2) invasive symptoms (e.g., repeat the reminder of the physiological response to trauma); (3) avoidance behavior (e.g., consciously avoid trauma-related things); (4) negative cognitive and emotional changes (e.g., the patient has a negative perception of the self and the world); (5) Changes in arousal and reactivity (e.g., aggressive behavior, inattentiveness and excessive vigilance about surroundings); (6) Symptoms severely affect an individual's function; (7) Symptoms must be last more than 1 month; and (8) exclude symptoms not caused by drug treatment, drug abuse, or other diseases. Studies have pointed out that PTSD has a heritability risk of up to 30-40%, indicating that the disease has genetic risk factors (Afifi et al., 2010). Many studies have proved that early stressors might cause long-term changes in rodent, primate, and human in multiple brain circuits and systems (Bremner, 2003; Chen & Baram, 2016; Danese & Baldwin, 2017; Koss & Gunnar, 2018). The HPA axis is an integrative system of controlling stress reactivity and stress hormone regulation in response to neuroendocrine stress. Many of its activities are well known and widely accepted. Tiwari and Gonzalez (2018) suggested that exposure to early life stress may sensitize the HPA axis responses, disrupting its normal function, and related cortisol production. It had been evidenced and supported by the results from animal studies (Arp et al., 2016; Hsiao et al., 2016; Bonapersona et al., 2019).. 3.

(21) Briefly, the HPA axis consists of three major components, including hypothalamus, anterior pituitary gland, and cortex of the adrenal gland. After exposure to physiological and psychological stressors, the afferent signal of the HPA axis through the amygdala is activated, leading to a cascade of hormone release (Tiwari & Gonzalez, 2018). The paraventricular nucleus (PVN) of the hypothalamus first releases corticotropin-releasing hormone (CRH). The circulation of CRH induces the anterior pituitary to release adrenocorticotropic hormone (ACTH), leading to the production and release of the glucocorticoid and cortisol from the adrenal cortex (Corcoran et al., 2003; Raabe & Spengler, 2013). The mineralocorticoid receptors (MRs) and glucocorticoid receptors (GRs) are two types of corticosterone (CORT) receptors. Both of them participate in the negative feedback regulation of the HPA axis. The corticosterone binding to hippocampal GRs regulates HPA axis functioning by inhibiting the further secretion of CRH from the hypothalamus and the subsequent release of cortisol (Dinkel et al., 2002; Chiba et al., 2012; Raabe & Spengler, 2013). Many studies have illustrated the debilitating effects and the role of childhood trauma in determining typical adult neuroendocrine regulation. Some of the studies proved that the PTSD and depression is closely related to early childhood trauma and neuroendocrine functioning. The gender differences in adult HPA axis response and related diseases from trauma in children may be a result of the vulnerability of gene-environmental the interactions regulate (Tiwari & Gonzalez, 2018). 4.

(22) Brain structures correlated to the stress-induced behavioral abnormalities During childhood, the critical brain contexture related to emotional memories including limbic system (hippocampus & amygdala) and prefrontal cortex are under-development, which making it vulnerable to the inadequate stress exposure. The damage to these structures will not only affect the normal function but also lead to mental illness in later life (Wei et al., 2015; Albrecht et al., 2017; Romeo, 2017). Some studies on adolescence have also shown that the over-reactive amygdala after trauma may trigger the vulnerability of PTSD (McLaughlin et al., 2014). In the group of adolescents with PTSD, the area of the prefrontal cortex that is responsible for emotion regulation is smaller than the control group (Keding & Herringa, 2015), and the area of the amygdala activated by emotion is activated by emotion larger (Weems et al., 2015). As described above, brain regions that trigger anxiety disorders and posttraumatic stress disorders such as the amygdala, hippocampus, and prefrontal cortex are also involved in the brain's emotional response for threats and fears regulation (Garza & Jovanovic, 2017). The sensory information of conditioned stimulation (CS) and unconditioned stimulation (US) are integrated into the amygdala. The amygdala is also the critical structure of regulating fear responses. These fear signals are transported to the lateral nucleus (LA) of the amygdala, and the central amygdala (medial [CeM] and lateral [CeL] subdivisions) is 5.

(23) responsible for sending out signals to the hypothalamus and brainstem structures. The intercalated cell mass (ITC) regulates the transmission of information between the basal nucleus (BA) and the central amygdala. The medial prefrontal cortex (mPFC) and hippocampus will mutually regulate the amygdala's signal output to the subcortical brain area and activate the fear reflex. As described above, brain regions that trigger anxiety disorders and post-traumatic stress disorders, such as the amygdala, hippocampus, and prefrontal cortex, are involved in the brain's emotional response that regulates threats and fears (Figure 1) (Ross et al., 2017).. Glucocorticoid effects on the function of the hippocampus Hippocampus plays an essential role in encoding, consolidation, and retrieval. It also implicated in spatial memory, sleep-dependent memory, and long-term memory of episodic and semantic (Bartsch, 2012). Therefore, it has become a core target in many human memory researches. At the network level, the neuroplasticity of the hippocampus circuit drives connectivity, structural changes, and behavioral outcome changes (Finke et al., 2013; Ryan et al., 2015). Stress may cause adaptability levels to multiple levels of maladaptive consequences in the body, and is thought to cause a variety of neuropsychiatric diseases. Among them, functional defects are observed in the hippocampus, such as cognitive impairment (Dohring et al., 2014; Lucassen et al., 2014). Numerous studies have shown that acute stress and glucocorticoids could damage the hippocampus of 6.

(24) rodents and humans, affecting spatial learning and memory retrieval (de Quervain et al., 1998; Morgan et al., 2006; Cazakoff et al., 2010; Olver et al., 2015). Under stressful conditions, glucocorticoid receptors would activate hippocampal function, leading to synaptic plasticity, neuronal survival, and hippocampal nerve development, i.e., acute behavioral stress appeared to affect the change of the hippocampal LTP and LTD in CA1 (Howland & Wang, 2008). Some studies have pointed out that the formation of LTP in animal models could enhance learning ability; while its weakening would retard learning. Clearly, LTP is associated with increased hippocampal synaptic transmission. Earlier studies, including ours (Matsuzaki et al., 2004; Wang et al., 2019). Ko et al. (2018) have demonstrated that the over phosphorylation of extracellular signal-regulated kinases (ERKs) in the hippocampus and the elevated hippocampal LTP response were correlated to the increase of depression-like animals in the neonatal dexamethasonetreated animals. The over phosphorylation of ERKs in the dorsal hippocampus, an essential structure for regulating the extinction of conditioned fear, had been reported to be a signal participating in the behavioral abnormalities. Cognitive impairments and LTP impairments are increasingly recognized as events accompanying and reflecting experimental depression, anxiety, and other experimental stress-related chronic psychological disorders. Therefore, disturbance in LTP may represent a reliable marker of synaptic plasticity impairment underlying depressive and anxiety states (Gulyaeva, 2016). 7.

(25) The related genes for stress-induced behavioral abnormalities related genes Brain-derived neurotrophic factor (BDNF) is expressed in human and the rodent brain abundantly. It can promote the growth and differentiation of neurons and regulate adult synaptic plasticity (Huang & Reichardt, 2001; Lu et al., 2014). The activity of BDNF is associated with many stressrelated mental illnesses, especially emotional disorders (Autry & Monteggia, 2012; Lu et al., 2014). The expression of BDNF is regulated by neuronal activity (Thoenen, 1991). Besides, by animal model, the role of BDNF in the regulation of the stress in the HPA axis has been demonstrated. More importantly, the activation of BDNF correlated to glucocorticoid activity has been investigated BDNF reveals a close relationship between glucocorticoid (Tapia-Arancibia et al., 2004; Taliaz et al., 2011; Jeanneteau et al., 2012; Alves et al., 2017). Many stress-related studies have shown that BDNF receptor TrkB signaling is mediated by glucocorticoids (Jeanneteau et al., 2008). Certain stress would stimulate the secretion of glucocorticoid (GC) and induce the expression of probrain-derived neurotrophic factor (pro-BDNF) and tPA protein. The produced could then tPA cleave plasminogen into plasmin, which in turn proteolytically processes pro-BDNF protein into BDNF. Mature BDNF then binds and activates TrkB receptor. TrkB phosphorylation will induce further phosphorylation and activate the Erk1/2 MAPK signaling pathway to mediate the enhancement of contextual fear memory (Revest et al., 2014), learning memory, and spatial memory (Miranda et al., 2019) (Figure 8.

(26) 2). The BDNF acts on specific receptors to regulate the physiology of neurons and glial cells (Williams & Umemori, 2014) associated anxietyrelated mental illness (Chen et al., 2006), such as depression (Monteggia et al., 2007) and PTSD (Roth et al., 2011; Zovkic et al., 2013). These changes are also crucial in the pathogenesis of affective disorders. In addition of the BDNF, the FK506-binding protein 5 (FKBP5) gene has also been widely studied. Previous studies also found that several genes, including brain-derived neurotrophic factor, FK506-binding protein 5 (FKBP5), participate in the long-term effect of stress on those subcortical nuclei. FKBP5 is a member of the 51-kDa immunophilin protein family. It can regulate immune response, protein folding, and translocation. FKBP5 is a functional regulator of GR signaling. The GR was dissociated from the FKBP5 / HSP70 / HSP90 complex and transferred to the nucleus through the interaction with FKBP4 and certain members of motor proteins, which promoted nuclear translocation and transcriptional regulation (Wochnik et al., 2005) (the detailed mechanism summarized in Figure 3). Studies have shown that increased single nucleotide polymorphisms (SNPs) in the FKBP5 locus can increase the risk of neuroendocrine stress disorders (Buchmann et al., 2014) and childhood-related abuse-related psychiatric disorders in adulthood (Appel et al., 2011; Klengel et al., 2013). There is evidence suggesting that single nucleotide polymorphism (SNPs) of FKBP5, rs1360780 interacts with stressors to elicit major depression (Matosin et al., 2018) and anxiety (Scheuer et al., 2016). Also, Matosin et al. (2018) proposed that early life stress-mediated increased expression of 9.

(27) FKBP5 epigenetics, which is related to functional changes in the amygdala and hippocampus.. The impact of stress on the excitatory neurotransmission The hippocampus is the brain region mainly involved in controlling learning and memory. The technique long-term potentiation (LTP), is frequently used to study the behaviors related to learning and memory (Malenka & Nicoll, 1999; Cooke & Bliss, 2006; Wang et al., 2019). In the process of inducing LTP, the postsynaptic NMDA receptors are activated, increasing intracellular calcium concentration and a series of Ca2+dependent signaling cascade (Malenka & Nicoll, 1999; Cormier et al., 2001; Myhrer, 2003; Baudry et al., 2015; Wang et al., 2019). The N-methyl-d-aspartate receptor (NMDAR) has been proved to affect emotion-related behaviors in rodent models and plays a vital role in many CNS diseases, including anxiety, depression, schizophrenia, and epilepsy (Delawary et al., 2010; Lang & Borgwardt, 2013; Paoletti et al., 2013; Ren et al., 2019). NMDARs are mainly a heteropolymer composed of the subunits NR1, NR2, and NR3. The NR2 subunit determines many characteristics and functions of NMDAR (Yang et al., 2017). In particular, the NR2A and NR2B subunits are the main subunits in cortical and hippocampal excitable pyramidal cells (Monyer et al., 1994; Wang et al., 2009). During the entire developmental process from birth to adulthood, the ratio of NR2A to NR2B changes dynamically. The performance of 10.

(28) NMDAR containing expression of NR2A relatively increases, while the performance of NMDAR containing NR2B expression relatively decreases (Carmignoto & Vicini, 1992; Monyer et al., 1994; Sheng et al., 1994; Wang et al., 2009). This change in the subunit composition has been postulated to be related to the increase of synaptic plasticity induction. Previous studies have found that NMDARs containing NR2A induces LTP in the hippocampal CA1 region, while NMDAR containing NR2B is individually responsible for inducing LTD (Liu et al., 2004; Kumar & Foster, 2019). A research review has shown that the calcium entering postsynaptic cells via NMDA receptors activates Calmodulin-dependent kinase II (CaMKII), in addition, protein kinase A (PKA) and Protein Kinase C (PKC) were also reported to modulate synaptic potentiation (Timmermans et al., 2013). Moreover, Whitehead et al. (2013) given animals acute restraint stress to examined hippocampal synaptic plasticity. They found the magnitude of LTP in excitatory postsynaptic potentials was significantly in stressed animals compared with control animals, and the LTP induced by HFS relies on the synaptic activation of NMDARs in the hippocampus.. Stress-induced abnormality of cation-coupled chloride transporters (CCCs) Several factors regulate the formation of long-term potentiation (LTP). (1) Activation of postsynaptic NMDA receptor (NMDAR) (Collingridge et al., 1983): The influxes Ca2+ leads to activation of calmodulin-dependent 11.

(29) Ca2+ /calmodulin-dependent protein kinase (CaMKII) (Lisman et al., 2012; Lisman, 2017). (2) Induction of LTP is facilitated by down-regulation of GABA receptor-mediated inhibition (Artola & Singer, 1987; Kotak et al., 2017). Gamma-aminobutyric acid (GABA) mediates the inhibition of most nervous systems in the brain through GABA-A receptors (GABAARs) (Olsen & Sieghart, 2009; Avoli & Krnjevic, 2016). GABARs are pentameric membrane proteins of the form of 2α, 2β, and 1γ (Patel et al., 2014). Many accumulated evidences suggest that GABA is a predominant inhibitory neurotransmitter in the brain and plays an essential role in controlling stress (Tsukahara et al., 2015). Stress can change the expression of GABA receptor subunits and glutamate decarboxylase (GAD) and promote GABA to produce defects (Peterlin et al., 2009; Kolata et al., 2018). Because the GABAA receptor can control the Cl− into the cell, and the activation has an indispensable relationship with the intracellular Cl− concentration ([Cl−]i) (Tsukahara et al., 2015). The GABA response modulates the surface expression of cation-chloride cotransporters (CCCs). The CCCs are transmembrane proteins that mediate the transport of chloride ions across the cell membrane (Blaesse et al., 2009; Ko et al., 2014). There are two main types of CCC that could facilitate the cotransportation of Cl− with sodium ions and/or potassium ions. One type of these, Na-K-2Cl cotransporters, including NKCC1 and NKCC2 are Na2+, K+, Cl− cotransporter. While the other type consists of four members include the KCC1, KCC2, KCC3 and KCC4 are in the K-Cl cotransporters (Mount et al., 1998; Kahle & Staley, 2008; Dzhala et al., 2010; Krystal et 12.

(30) al., 2012). NKCC1 is distributed throughout the body, moving chloride ions into the cell and increases the concentration of chloride ions. This would promote excitatory activation of GABAAR (Haas, 1994). KCC2 is expressed throughout the central nervous system (CNS), and mainly transports chloride ions out of the cell, maintaining low [Cl−]i. It is important for GABAergic inhibitory signaling (Rivera et al., 2005; Vinay & Jean-Xavier, 2008) (Figure 4). The NKCC1 and KCC2 are Cl− transporters that control intracellular chloride homeostasis in neurons (Tzanoulinou et al., 2014). An earlier study showed that certain chronic stress could lead to a decrease in the expression of KCC2 but overexpression of NKCC1 in the cell membrane. It seemed that the KCC2 / NKCC1 balance associated stress-induced behavioral disorders is disrupted, and promote the inflow of chloride ions into the cell leading to an over-excitatory of the cell (Tsukahara et al., 2015). The drugs for inhibitory neurotransmission are believed to be an effective treatment of anxiety disorders. For instance, benzodiazepines are specific to modulate GABAergic neurotransmission. GABA receptor agonists and anxiety-associated in the molecular components of inhibitory synapses have drawn much of the attracted attention. Loop diuretics, such as furosemide and bumetanide, are considered to elicit anxiolytic effects, based on the tests in rodent models by fear conditioning (Krystal et al., 2012). Both furosemide and bumetanide have a higher affinity for NKCC1 than KCC2. Furosemide inhibits the NKCC symporter and acts as a noncompetitive antagonist at GABAA receptors (Gutschmidt et al., 1999). 13.

(31) Bumetanide is an NKCC1 specific inhibitor, exhibiting a low affinity with KCC2 (Blaesse et al., 2009). It can mediate anxiolytic effect by antagonizing NKCC1 but not KCC2, and used to treat aberrant NKCC1 expression related diseases. Ko et al. (2014) used the inhibitory avoidance test, extracellular recordings, and western blots to assess the role of NKCC1 in hippocampal function. The results showed that intravenous bumetanide 30 minutes before the learning phase prevented inhibitory avoidance learning. Bumetanide does not affect the spontaneous movement of animals. They demonstrated the use of pharmacological intervention through bumetanide's NKCC1 inhibitor to regulate the hippocampus.. Research aim and significance Mounting shreds of evidence suggest that acute stress would cause the release of glucocorticoids and the changes of glutamate neurotransmission in the hippocampus and the amygdala. Also, many studies also showed that the dysfunction of glutamatergic neurotransmission is one of the features of stress-related mental illness (Popoli et al., 2011). Results obtained from the present study will expand our knowledge on the long-term adverse effect of juvenile maltreatment on the hippocampal function. It might also provide some preliminary data for examining the possibility of using NKCC1 inhibitor as a therapeutic agent for cuing those behavioral and functional abnormalities. 14.

(32) MATERIALS AND METHODS Animals Four weeks old male C57BL/6 mice were purchased from BioLASCO Taiwan Co., Ltd. Mice were housed in the animal facility of the National Taiwan Normal University. The diurnal cycle began from 8:00 and ended at 20:00. Animals were allowed the access to water and food ad libitum at all times. All behavioral procedures were performed in the afternoon (from 13:00 to 17:00). All experimental procedures adopted were approved by the Laboratory Animal Management Committee of National Taiwan Normal University and compliance with the Guide for the Care and Use of Laboratory Animals of the Council of Agriculture, Republic of China.. Juvenile immobilization treatment (J_IMO) The mice were given a two-hour for once to immobilization stress treatment (IMO) at the postnatal day 35 and 36 (PND). The IMO procedure adopted in this study is similar to that was reported by Lin (2019, dissertation). Juvenile mice were placed into a triangular restraint bag with a few seams to ensure that the animals breathe smoothly. To prevent them escape from the restraint bag, a cone-shape cream nozzle was attached to the head of the mice. All these procedures were only be performed in the afternoon, and room temperature were maintained at 25 °C. Animals were subjected to two consecutive days IMO, depending on the purpose of the experiment. 15.

(33) Inhibitory avoidance test (IA) The inhibitory avoidance test is widely used to study the hippocampal dependent learning and memory. During experimental training, mice are placed in a behavior box. The behavioral chamber consists of a light compartment and a dark compartment separated by a sliding door in the middle. A single training was given, animals were then returned to the housing cage and subjected to the testing twenty-four hours later. The time (latency) of the mice from the light compartment move into the dark compartment was served an index of avoidance learning (Figure 5). The general behavioral procedure of IA is composed of four different phases (acclimation, training, testing, and extinction). The details are summarized below: 1. Acclimation It is aimed to reduce the animal's context fear to the behavioral apparatus and working procedure. During the acclimation, animals were placed in the light room for 40 seconds, and subsequently the sliding door is opened. Once the animal passes from the light room to the dark room, the door were closed, and the animals were returned to the breeding box after 2 minutes. 2. Training Twenty-four hours after the “acclimation”, the animals were subjected to test training. Again, animals were placed in the light room for 40 seconds, and the separating door is opened. After entering the dark room, the door 16.

(34) is immediately closed and animals were given a footshock (0.8 mA, 2s) two seconds later. Record the incubation period of animals from the bright room into the dark room. 3. Testing Twenty-four hours after the “training” the animals were returned to light room. They were placed in the light room for 180 seconds before the door opening. The escape latency was recorded as an index of avoidance learning. In addition, the recording was stopped if the animal stays in the light compartment for more than 10 mins. 4. Extinction Twenty-four hours after the “testing”, animals were returned to light compartment of the behavioral chamber. The procedure is similar to the training except that no foot shock was given.. Open field test (OFT) The mice were placed in the middle of a test chamber with diameters of 40 x 40 x 30 cm (length x width x height). The floor of the chamber was divided into sixteen equal sized square zones. The middle four-square zones were defined as a center area, and the rest twelve square zones, surrounding the center area, were defined as a peripheral area (Figure 6). The animal’s horizontal moving activity, the number of crossing and percentage of time spent between the center and peripheral areas were 17.

(35) repetitively monitored by a digital camera for a total of 10 minutes. Rodents generally tend to move around. If the frequency and time in the central part of the venue increase, it means that the degree of anxiety is reduced, and the degree of anxiety in experimental animals is evaluated. The data were analyzed by a commercial tracking software SMART VIDEO TRACKING (manufactured by Panlab Harvard Apparatus, Spain). The illumination of the behavior room was held on 30 Lux. After each test, the chamber was thoroughly washed with 70% ethanol. The OFT procedure was performed similar to the previous study by Lin (2019, dissertation).. Elevated plus-maze (EPM) The maze was conducted experiment under glimmer (elevated 30 lux). The maze itself were constructed of two open (30 × 5 cm) and two closed arms (30 × 5 cm) by a 14 cm height walls. Closed arms possessed black walls 40 cm in height, and all arms were connected by a central 5 cm by 5 cm square section (Figure 7). Through the rodent's aversion to open arms that are hanging high and empty. If the animal enters the open arm for a longer time and more frequently, it means that the degree of anxiety decreases. The observer was situated in a neighboring room, and each session of the test was recorded by video (Smart 3.0) for subsequent analysis. Mice were individually placed in the center of the maze facing a closed arm and 18.

(36) allowed 5 minutes of free roaming. Metrics will include number of entries into an open arm, number of entries into a closed arm, total time spent in open arms, and total time spent in closed arms. The EPM procedure we used to be a modified one derived from our previous study (Lin, 2019, Master Dissertation).. Brain slice extracellular recording Coronal slices of mice brain (300 μm thick) were prepared with a double-edge blade. The slices were placed in oxygenated artificial cerebrospinal fluid (ACSF) for at least 1 h in room temperature before recording. Each slice was then transferred to a recording chamber, where it was held on between two nylon nets and maintained at 32 ± l ℃. The chamber consisted of a circular well with a volume of 1~1.5 ml and was perfused with ACSF constantly at the rate of 3-4 ml/min. A bipolar stimulating electrode (SNE-2OOX, Kopf Instrument, USA) was used in this study. Field excitatory postsynaptic potential (fEPSP) was recorded extracellularly by using a glass microelectrode filled with 3M of NaCl (38 MΩ). The stimulus electrode was placed in the CA3 region of hippocampus and the recoding electrode was set to the CA1 region of the hippocampus, respectively. The evoked fEPSP signals were then recorded by an Axoclamp-2B amplifier (Axon Instruments, USA). The responses were elicited by low square-wave pulses delivered at a 20 s interval, filtered at 1 kHz and digitized at 5 kHz (Digidata 1322A; Axon Instruments, 19.

(37) USA). The stimulation voltage was adjusted individually for each trial to produce fEPSP, by which 30-40% of the maximal response could be evoked. The strength of synaptic transmission was quantified by measuring the initial slope of the fEPSP, and the results were analyzed by pCLAMP software (Version 10.2; Axon Instruments, USA). High frequency stimulation (HFS) were initiated as LTP by test pulse intensity (1-second pulse of 100-Hz stimuli separated by an interval of 20 seconds). The input-output (I/O) curve and paired-pulse facilitation (PPF) experiment were performed with separate brain slices. I/O relationship and PPF were assessed to indicate the basal synaptic transmission functions and presynaptic plasticity. The I/O curve was recorded and calculated with incremental stimulation intensities of 65, 70, 75, 80, 85 mV. The PPF test delivered with interpulse intervals of 20, 50, 100, 150 and 200 milliseconds.. Administration of bumetanide and furosemide For extracellular recording, bumetanide and furosemide were first dissolved in 100% DMSO to make a 10 mM and 50 mM stock solutions, respectively. The stock bumetanide was diluted to 5 μM and 10 μM by artificial cerebrospinal fluid (aCSF), and the shock furosemide was diluted to 50 μM by aCSF. The bumetanide dosage used were based on previous studies (Ko et al., 2014; Ko et al., 2018). The final concentration of DMSO was approximately 0.05 ~ 0.10%. During the experiment, the hippocampal slices were first perfused with 20.

(38) pure aCSF, and then perfused with aCSF containing bumetanide and furosemide. The baseline signal was recorded for 20 minutes (only recorded aCSF in the first 10 minutes, suprafused bumetanide and furosemide in the last 10 mins). After the LTP induction, slices were suprafused with bumetanide or furosemide for 10 minutes. Then followed with a 50 mins recording with the continues suprafusion of aCSF.. Real-time polymerase chain reaction The mice were sacrificed at PSD-42 and their brains were frozen at 70℃. The dorsal hippocampus (dHip) (bregma -1.34~ -2.06 mm, X: 2.5 ~ 2.5 mm, Y: 1.5~2.5 mm) regions of the brain were punched out using a 1 mm diameter Integra Miltex Disposable Biopsy Punches (skin biopsy punch). The collected tissues were stored at −70°C for qPCR analysis. From these tissues, total RNA was extracted by using a LabDrepTM RNA Plus mini kit (cat. No. LPRS100, TW) by using the provided protocol. A Nanodrop 1000 spectrophotometer were used to measure the concentration of the obtained RNA. Then configure the same ng of RNA (total 13 µL), mix 1 µL 10 µM oligo dT. Incubate for 10 minutes at 70 °C, and immediately treat it at low temperature (incubated on ice). To this solution, 4 µL of 5x RT buffer (250 mM TrisHCl, pH 8.3, 375 mM KCl, 15 mM MgCl2, 50 mM DTT), 1 µL of 10 mM dNTP, 1 µL EasyScriptTM III RTase (200 units, Bioman, Taiwan) were added, and the reverse transcription was performed at 60 ℃ for 2 hours. 21.

(39) The obtained cDANs were then subjected to quantitative PCR by using PowerUp SYBR Green Master Mix (Cat. A25742, ThermoFisher, USA) and quantified by StepOnePlus Real-Time PCR Systems (ThermoFisher, USA). Specific primers for PCR were designed and used in the expression analysis of Fkbp5, Gabra1, Grin2a, Slc12a2, Slc12a5 and Gapdh (reference gene). The relative level of each gene expression was normalized with the expression levels of Gapdh. Graphics data were represented by fold change obtained by the 2-△△Ct method. These experiments were collaborated with Mr. Lin, Wei-Hsing.. Appendix primer sequences for qPCR Gapdh Product size. 129. Reference Sequence. Forward. GTGGACCTCATGGCCTACAT. Reverse. TGTGAGGGAGATGCTCAGTG. NM_008084.3. Fkbp5 Product size. 163. Reference Sequence. Forward. GGAGCCGACTGTGTGTGTAA. Reverse. CAGTCTCCTTGGCCCACAAT. 22. NM_010220.4.

(40) Gabra1 (GABAA1 receptor gene) Product size. 138. Reference Sequence. Forward. GAACAGTTCCTGCTGACTCC. Reverse. CTCTGGAAAGCGAGACATGC. NM_001359035.1. Grin2a (NR2A receptor gene) Product size. 207. Reference Sequence. Forward. CTGCCTTGTGGTCCTCAATC. Reverse. GCCCTCAATCACTCTGACAC. NM_008170.4. Slc12a2 (NKCC1 gene) Product size. 134. Reference Sequence. Forward. CACAGTGAATCTCGATGCAC. Reverse. CTTGAGACTGTTTGACCAGG. NM_009194.3. Slc12a5 (KCC2 gene) Product size. 146. Reference Sequence. Forward. CAGTGGTTTTGCCTTTTGGG. Reverse. GTGGGCTGTTTTCATCAACG. 23. NM_020333.2.

(41) Gapdh: Glyceraldehyde 3-phosphate dehydrogenase FkbpP5: FK506 binding protein 5 Gabra1: Gamma-aminobutyric acid (GABA) A receptor, subunit alpha 1 Grin2a: Glutamate receptor, ionotropic, N-methyl-D-aspartate epsilon 1 (NMDA2A) Slc12a2: Na-K-Cl cotransporter 1 (solute carrier family 12, member 2) Slc12a5: K-Cl transporter member 5 (solute carrier family 12, member 5). Western blot The mice were sacrificed at PSD-42 and their brains were frozen at −70℃. The dorsal hippocampus (dHip) (bregma -1.34~ -2.06 mm, X: 2.5 ~ 2.5 mm, Y: 1.5~2.5 mm) regions of the brain were punched out using a 1 mm diameter Integra Miltex Disposable Biopsy Punches (skin biopsy punch). The obtained tissues were then subjected to western blot analysis. We used the T-PER Tissue Protein Extraction Reagent (Thermo scientific, USA) with the HaltTM Protease & Phosphatase Single-Use Inhibitor Cocktail (Thermo scientific, USA), to homogenize tissues by ultrasonic homogenization. Samples were centrifuged at full speed (16,000g) for 10 minutes at 4°C and the precipitate was discarded. Total protein concentration was determined by using a Bio-Rad Bradford protein Assay Kit (Bio-Rad, Hercules, USA). Protein from each sample (30 μg each) were electrophoresed on SDS PAGE, and the resolved protein were 24.

(42) electrically transferred to a 0.45 µm PVDF membrane (Merck Millipore, Germany) with 70V for 2 hours. The membrane was blocked in TBST (0.1% Tween-20) containing 5% nonfat milk for 1 hour at room temperature. Primary antibodies used were specifically against NKCC1 (1:2000 in 5% nonfat milk, 14581s, Cell Signaling Technology, USA) and beta actin (1:5000 in 5% BSA, Ab8227, Abcam, UK). The PVDF membranes were shake in a primary antibody solution overnight at 4°C. Next day, pour out the antibody solution and store at 4°C. The PVDF membrane was washed with TBST for five min, a total of three times. Then add the HRP-conjugated secondary antibody (anti-rabbit IgG, HRP-linked antibody, 1:3000 in 5% nonfat milk, 7074s, Cell Signaling Technology, USA) and shake for 1 hour at room temperature. After pour out the secondary antibody, the PVDF membrane was washed with TBST for five minutes, a total of three times. The detection reagent was used by Ex-CL Western Chemiluminescent Kit (Cat# W-3408-0, Goal Bio, Taiwan) and were detected by LAS 4000 (GE Healthcare, USA). Evidence of protein presence were analyzed by image studio lite (ver 5.2) analysis software. These experiments were collaborated with Mr. Lin, Wei-Hsing.. Statistics All the calculated data were presented as mean ± S.E.M. The data of inhibitory avoidance testing results were analyzed by ANOVA two-way analysis, Multiple comparisons, and Sidak’s multiple comparisons test of 25.

(43) variance. Whereas, that of behavior testing (open field test and elevated plus-maze) results were analysis by using a one-way ANOVA. A two-way ANOVA was used when there was more than one independent variable. The statistical analysis on LTP, I/O and PPF experiments were performed by using the two-way ANOVA and t-test. Real-time PCR data were analyzed using Mann-Whitney test. Protein level data in western blot were analyzed with an unpaired Mann-Whitney test. Probability levels of less than 0.05 (p <0.05) were considered to be significant. All data were analyzed with GraphPad Prism software.. 26.

(44) RESULTS Experiment-1: Examine the juvenile immobilization treatment effects on the inhibitory avoidance learning. Rationale: Previous animal studies have shown that adverse early life experiences would lead to long-term effects on brain function, cognition, emotional development, learning memory, and increase anxiety-like behavior (Kosten et al., 2012; Teicher & Samson, 2016; Loi et al., 2017). These all potentially might increase the risk of developing stress-related psychopathology in later life (Kendler et al., 2000). Therefore, we first decided to evaluated the effect of immobilization treatment (IMO) by the inhibitory avoidance learning response of male juvenile mice.. Procedure: Only C57BL/6 male mice were chosen for avoiding the possible confounded effects of estrus cycle on the emotional behavior. Briefly, juvenile mice (five weeks old) were assigned to two groups, control group and two days immobilization treatment group (J_IMO group). Animals received the J_IMO treatment were at the age of PND-35 &36. Along with control group, they were then subjected to the inhibitory avoidance test (IA) at the age of PND-42. The detailed testing procedures were described in the materials and methods session. The latency of the animals entering the dark compartment after opening of the sliding door were recorded as the 27.

(45) escape latency and served as an index of avoidance learning.. Results: Our results showed that there was a significant increase of the escape latency found between the training and testing session of both control and J_IMO groups (two-way ANOVA F (1, 25) = 12.83 and p = 0.0014). These suggested that both groups had been successful trained and exhibited the avoidance response to the footshock. The learned avoidance response was successfully declined in the control group after two times of extinction training (Ext-1 & Ext-2) (two-way ANOVA, Sidak's multiple comparisons test: p < 0.0001 in Ext-1 by control vs. J_IMO; p = 0.0171 in Ext-2 by control vs. J_IMO). In contrast, a significant blockage of extinction learning was observed in the J_IMO group. After additional two times of extinction training (Ext-3 & re-test), the J_IMO animals showed a decline in the learnt avoidance response and did not reach a significant level (p = 0.2102 in Ext-3 by control vs. J_IMO; p = 0.4348 in re-test by control vs. J_IMO). These results suggested an extinction learning defect occurred to the J_IMO treated animals (Figure 8A). Compare the components separately in Figure 8B. In the control group, p = 0.0053 by test compared with the corresponding train; p = 0.0172 by Ext-2 compared with the corresponding test; p = 0.009 by Ext-3 compared with the corresponding test; p = 0.0105 by re-test compared with the corresponding test (two-way ANOVA, Sidak's multiple comparisons test). In the J_IMO group, p < 28.

(46) 0.0001 by test compared with the corresponding train; p < 0.0001 by Ext1 compared with the corresponding train; p = 0.0061 by Ext-2 compared with the corresponding train; p = 0.0014 by Ext-3 compared with test; p = 0.0003 by re-test compared with the corresponding test (two-way ANOVA, Sidak's multiple comparisons test).. 29.

(47) Experiment-2: Evaluation of the juvenile immobilization treatment induced anxiety-like behavior in the male mice by open field test and elevated plus-maze test. Rationale: In our previous studies, we have demonstrated that IMO treatment could induce anxiety-like behavior in the adult male mice (Tu et al., 2019). In this study, we would also like to determine whether the J_IMO could also elicit a similar effect. This would reparent a fundamental step to the subsequent experiments.. Procedure: Additional control group and J_IMO animals were prepared and subjected to the open field test (OFT) and elevated plus-maze test (EPM) one week after the J_IMO treatment. In OFT, the number of times mouse entries to the center zone, the time spent in the center zone, and the total moving distance were recorded and analyzed. The same animals were then subjected to the EPM 24 hours later. The number of entries in each zone, the percent time spent in each zone, and the total moving distance were recorded and analyzed. For details, please refer to the materials and methods session.. 30.

(48) Results: There was no significant difference in the total moving distance of OFT and EPM among groups (Figure 9C, F). In the OFT test, the number of entries into the center zone and the time spent in the center zone were significantly decreased in the J_IMO treated mice group (Mann-Whitney test, the number of entries into the center zone: p = 0.0238 compared with the corresponding control group; the time spent in the center zone: p = 0. 0052 compared with the corresponding control group) (Figure 9A, B). These results suggested an increase of anxiety-like behavior existed in the J_IMO treated animals. In the EPM test, no significant difference was found on the number of entries and time spent in open arms among the control group and the J_IMO group (Figure 9D, E). In summary, we found that anxiety-like behavior is elevated in the J_IMO groups which could be detected by OFT but not by EPM.. 31.

(49) Experiment-3: Determine the hippocampal long-term potentiation of the IMO animals using brain slice extracellular recording. Rationale: Our data showed learnt avoidance response was enhanced and extinction was impaired in the J_IMO treated mice. We speculated that the neurotransmission of hippocampus might also be modified. Therefore, we examined this possibility here by in vitro extracellular recording.. Procedure: Additional control and J_IMO animals were prepared and sacrificed at the age of 6-week-old for the in vitro extracellular recording of the highfrequency (HFS) induced hippocampal long-term potentiation (LTP). The stimulation electrode and recording electrode were placed in CA3 and CA1 of hippocampus, respectively. The detailed experimental procedure is described in the materials and methods.. Results: The HFS-induced hippocampal LTP was significantly enhanced in the J_IMO group animals as compared with the corresponding control group, as the measurement was made 60 min after a tetanic stimulus (two-way ANOVA F (1, 12) = 6.455, p = 0.0259) (Figure 10A). The average magnitude of potentiation in the J_IMO group measured 10 min before 32.

(50) HFS was 98.7 ± 2.36 % and the control group was 103 ± 4 %. No significant differences were found (Mann-Whitney test, p = 0.4557). The average magnitude of potentiation in the J_IMO group, measured 40 min after HFS, was 316 ± 37.5 % and the control group was 103 ± 4 %. It was increased in J_IMO mice significantly (Mann-Whitney test, p = 0.0006) (Figure 10B). These results proved that IMO might provoke a long-lasting effect on the neurotransmission of hippocampus.. 33.

(51) Experiment-4: Determine the input-output curve and paired-pulse facilitation of the brain slices of J_IMO treated animals. Rationale: Our data showed J_IMO treated mice actually exhibited a long-lasting effect on the neurotransmission of the hippocampus. Thus, we would like to focus our interest in determining whether a pre-synaptic or post-synaptic mechanism is involved in such process.. Procedure: Animals subjected to J_IMO or control without treatment, were used to undergo in vitro extracellular recording at the age of six weeks. For the electrophysiological recording, a biphasic stimulation was used in all the experiments. The input-output (I/O) curve and paired-pulse facilitation (PPF) experiment were performed in separate brain slices. The stimulation electrode and recording electrode were placed in CA3 and CA1 of hippocampus, respectively. The I/O curve ratio was induced and calculated with incremental stimulation intensities of 65, 70, 75, 80, 85 mV. The PPF test delivered with interpulse intervals of 20, 50, 100, 150 and 200 milliseconds.. Results: The fEPSP slopes increased along with the incremental stimulation 34.

(52) intensities in the I/O curve (Figure 11A). There was a significant difference, between the control and J_IMO groups, in the trend of the slope increase (two-way ANOVA, F (1, 12) = 10.43, p = 0.0072). And, there was significant deviation, between the control and J_IMO groups in the slopes to fourth and fifth stimulation intensity density (Row 4, p = 0.0494; Row 5, p = 0.0182). We also carried out PPF test to verify the involvement of presynaptic mechanism. Results showed in Figure 11B, pairs of presynaptic fiber stimulation pulses delivered with interpulse intervals of 20, 50, 100, 150 and 200 milliseconds evoked nearly identical amount of PPF ratio on the slices from the control and J_IMO treated groups (two-way ANOVA, F (1, 60) = 0.1051, p = 0.747). These results suggested the hippocampal postsynaptic basal neurotransmission rather than presynaptic function was altered in the Schaffer collateral pathways of the J_IMO treated mice.. 35.

(53) Experiment-5: Examine the expression of HPA axis-related and neurotransmitter receptor genes of the hippocampal on the juvenile IMO animals by using real-time polymerase chain reaction. Rationale: Previous studies have shown that the Fkbp5 gene is a stress-related gene and regulates the HPA axis (Appel et al., 2011; Klengel et al., 2013). Some studies also pointed out that the Grin2a (NR2A receptor gene) expression regulation in the hippocampus is correlated to the developmental changes in cognition and synaptic function (Dumas, 2005; Kumar & Foster, 2019). MacKenzie and Maguire (2015) reported that repeated stress might induce a downregulation of Slc12a5 (KCC2 gene) and an upregulation of Slc12a2 (NKCC1 gene) in hippocampal cell membranes, such alterations in gene expression were found to modulate the Gabra1 (GABAA receptor gene) response to depolarization of hippocampal cells of adult male mice. The current experiment was aimed to analyze the possible IMO treatment associated the HPA axis regulation on the stress correlation genes Fkbp5, neurotransmitter receptor genes (Grin2a & Gabra1) and the genes coding for cation-coupled chloride transporters (Slc12a2 & Slc12a5) of male juvenile mice.. 36.

(54) Procedure: Separated groups of control and J_IMO animals were sacrificed at the age of six weeks. The brain tissues were collected and subjected to realtime PCR. The target gene expression was studied by real-time PCR using 2-△△Ct method (For details, please refer to the materials and methods session).. Results: According to the results of qPCR, we have identified the genes that were significantly affected by J_IMO. On the dorsal hippocampus, the expression of Grin2a (NR2A receptor gene) (Mann-Whitney test, two tailed, p = 0.0305) and Slc12a2 (NKCC1 gene) (Mann-Whitney test, two tailed, p = 0.0289) were significantly increased in the J_IMO (Figure 12A ~ E). However, no significant differences in Fkbp5 (Mann-Whitney test, two tailed, p = 0.4805), Gabra1 (GABAA receptor gene) (Mann-Whitney test, two tailed, p = 0.2646) and Slc12a5 (KCC2 gene) (Mann-Whitney test, two tailed, p = 0.1893) gene expression. The neurotransmitter genes result the validates found out in vitro extracellular recording research. The results showed that neurotransmitter receptor genes of Grin2a and the cation-coupled chloride transporters expression of Slc12a2 were upregulated increases in stress treated mice.. 37.

(55) Experiment-6: Analyze the hippocampal NKCC1 expression of J_IMO mice by Western blotting. Rationale: Our qPCR results have demonstrated an increase of hippocampal NKCC1 expression in J_IMO treated mice. This might provide some clue for the behavioral and electrophysiological changes resulted from juvenile immobilization. Here, we decided to verify the qPCR results by using Western blotting.. Procedure: The untreated control group and J_IMO animals were prepared and sacrificed at the PND-42. The hippocampus tissues were collected and the total proteins were used in Western blotting analysis.. Results: As compared with that from the control group, NKCC1 total cellular proteins was significantly elevated in hippocampus after IMO-treated (Mann-Whitney test, two tailed, p = 0.0411) (Figure 13). These results were rather consistent with to the qPCR data, implying that the activity of hippocampal NKCC1 is highly correlated to the behavior abnormalities in J_IMO mice.. 38.

(56) Experiment-7: The enhanced hippocampal HFS-LTP of the J_IMO animals were reversed by suprafusion of NKCC1 antagonist in a dosedependent manner. Rationale: As the results of qPCR and Western blotting both suggested hippocampal NKCC1 might participate in the J_IMO induced behavioral abnormalities. In another words, blocking the activity of NKCC1 should be able to alleviate the abnormalities. Therefore, we again used extracellular recording together with the administration of furosemide and bumetanide (NKCC1 antagonist) to test this hypothesis.. Procedure: The procedure was similar to that described in experiment-3 except that different doses of bumetanide were suprafused 15 minutes prior to and 15 min after the HFS. The animals were assigned to three groups as below in a dose‐ dependent test. They were all J_IMO un-treated mice: Age 6 weeks old (juvenile). Treatment. Number of animals. 0.1% DMSO. 6. 5 µM bumetanide. 6. 10 µM bumetanide. 6. 39.

(57) The animals were assigned to six groups as below the suprafusion of bumetanide blocked the formation of LTP in the IMO-treated: Treatment. Number of animals. Control. 6. J_IMO. 6. 6 weeks old. Control + 5 µM bumetanide. 6. (juvenile). J_IMO + 5 µM bumetanide. 6. Control + 50 µM furosemide. 6. J_IMO + 50 µM furosemide. 6. Age. Results: The obtained results showed that the J_IMO treated mice with suprafusion of 10 μM bumetanide could effectively block hippocampus LTP formation as comparing with that of 5 μM bumetanide group. The three groups exhibit no significant difference in HFS, but the column Factor was F (2, 15) = 3.344, p = 0.063 (two-way ANOVA) (Figure 14A). Therefore, we compare 5µM and 10µM bumetanide separately. The average magnitude of potentiation in the 0.1 % DMSO group measured 40 min before HFS was 176.1 ± 17.12 %, the 5 µM bumetanide group was 173.7 ± 34.82 % (Mann-Whitney test, p = 0.8182) (Figure 14B). They were no significant difference. The 0.1 % DMSO compared with 10 µM bumetanide. The average magnitude of potentiation in the 10 µM 40.

(58) bumetanide measured 40 min before HFS was 115.3 ± 3.426 % (MannWhitney test, p = 0.0087) (Figure 14C). The 10 µM bumetanide wase significant difference decreased. J_IMO treatment effect on the high-frequency stimulation-induced hippocampal long-term potentiation enhanced in the juvenile mice can used the suprafusion of furosemide and bumetanide blocked the formation of LTP in the hippocampus, in a dose‐dependent manner. The HFS-induced hippocampal LTP was significantly enhanced in the J_IMO group animals as compared with that in the corresponding control group (Figure 14D), the J_IMO group measured 10 min before HFS was 99.91 ± 2.224 % and control group was 85.33 ± 2.658 %. The average magnitude of potentiation in the J_IMO group measured 40 min before HFS was 329.9 ± 38.31 %, the control group was 157.8 ± 17.12 %, and we found J_IMO increased significantly (two-way ANOVA, F (5, 30) = 9.619, P<0.0001) (Figure 14D). Suprafusion of NKCC1 antagonist furosemide and bumetanide in 50 µM and 5µM was restored the HFS-induced fEPSP in the hippocampus LTP formation in J_IMO male mice (Figure 14D). Therefore, J_IMO compared with 5µM bumetanide and 50µM furosemide separately. The average magnitude of potentiation in the control + 5 µM bumetanide measured 40 min before HFS was 142.1 ± 15.14 %, J_IMO + 5 µM bumetanide was 149.5 ± 9.954 % (Kruskal-Wallis test, p = 0.004) (Figure 14E). The average magnitude of potentiation in the control + 50 41.

(59) µM furosemide group measured 40 min before HFS was 174.2 ± 17.56 %, and J_IMO + 50 µM furosemide group was 169.6 ± 18.31 % (KruskalWallis test, p = 0.0062) (Figure 14F).. 42.

(60) DISCUSSION Summary of the results In the present study, we explored the neural mechanism of the longterm adverse behavioral effect by mimicking the adolescent abasements using a modified rodent immobilization (IMO) model. Briefly, juvenile mice at the postnatal day 35 and 36 (PND-35 & PND-36) received the juvenile immobilization treatment group (J_IMO). One week later (PND42), animals were subjected to the behavioral, electrophysiological, and biochemical experiments. We compared the differences in the learning memory and anxiety-like behavior between the control and the J_IMOtreated groups of mice. Behavioral experiments conducted in this study included inhibitory avoidance task (IA), open field test (OFT), and elevated plus maze (EPM). IA was carried out to evaluate the possible defect in learning associated with J_IMO. Both OFT and EPM were applied to analyze the J_IMO treatment-induced anxiety-like behavior. Besides, we also used in vitro extracellular recording, real-time polymerase chain reaction (qPCR), and western blot (WB) to investigate the long-term effect of J_IMO treatment on hippocampal function. The dorsal hippocampus tissues (DH) were collected at PND-42 and analyzed by qPCR or WB for the molecular mechanism of J_IMO. We also examined the rescue effect of bumetanide (a loop diuretic) on the uncontrolled highfrequency stimulation-induced long-term potentiation (HFS-LTP) in J_IMO treated mice. Our previous results demonstrated that anxiety-like. 43.

(61) behavior was long-term increased in the IMO-treated male mice. We found that both juvenile (five-weeks old) and adult (eleven weeks old) IMOtreated mice showed impairment on the extinction of conditioned fear in adulthood, and revealed an elevated hippocampal HFS-LTP response. These results suggested that the impact of J_IMO treatment might persist throughout adulthood (Leo, 2019, dissertation). The results of the present study showed that two times of immobilization treatment, with twenty-four hours apart, at the age of five weeks old resulted in an interference with the extinction of inhibitory avoidance task (Figure 8). These indicated the impact of J_IMO is not restrained in the extinction of fear memory but also in the extinction of inhibitory avoidance memory. Unlike Lin’s results (Lin, 2019, dissertation), our OFT results indicated an increase of anxiety-like behaviors in the J_IMO animals which suggesting the juvenile mice were more vulnerable to the immobilization stress (Figure 9A). As shown in Figure 10, the hippocampal HFS-LTP elevated significantly in the brain slices of J_IMO-treated mice. This could be triggered postsynaptic ally or presynaptic ally. Interestingly, we detected an upward fEPSP ratio in the input-output curve (I/O curve) (Figure 11A), but no change in the ratio of pair-pulse facilitation (PPF) (Figure 12B). These results implied that the elevation of the hippocampal HFS-LTP might have resulted from a postsynaptic mechanism. Result of the qPCR showed that the hippocampal expression of Grin2a 44.