早產兒產後使用糖皮質類固醇Dexamethasone 對神經發育之影響 --從臨床問題到動物模型探討

Effects of postnatal dexamethasone therapy on neurological development in premature infants– from clinical question to animal model

蔡政憲(Cheng-Hsien Tsai)

指導教授 ： 謝正勇

共同指導教授 ： 林其和 ； 簡基憲(Chi-Hsien Chien)

國立臺灣大學/醫學院/臨床醫學研究所/碩士(2004年)

https://doi.org/10.6342/NTU.2004.01908

摘要

Dexamethasone (Dex)是一種人工合成的糖皮質類固醇，強效而且持久。它的藥理機制為抑制早期和晚期的發炎反應並且誘導細胞色素P450相關酵素的形成。在臨床上慢性肺疾病是早產兒重要的致死因和致病因，Dex經常用為治療和預防早產兒慢性肺疾病。
為了預防慢性肺疾病而給與Dex治療的早產兒在文獻上報告有生長遲緩和神經發展落後的不良影響，我們台灣的早產兒聯合研究團隊，今年報告在出生12小時內開始給予Dex治療的隨機指派、雙盲研究，在學齡兒童時期的發展有顯著的生長遲緩、神經動作不良、認知功能發展落後、和顯著障礙的比率上升等不良作用。但是我們對Dex的不良作用機制仍然尚未清楚。
海馬迴結構屬於邊緣系統，含有高量的糖皮質類固醇接受器，是腦下垂體—下視丘—腎上腺軸的主要抑制來源，它的功能包含認知發展、下視丘控制、情緒表達和學習記憶，海馬迴結構在胚胎發育時期是最早出現糖皮質類固醇接受器的腦部組織。從臨床研究的觀察，我們的假設是產後早產兒接受外來性糖皮質類固醇藥物的曝露影響到海馬迴結構的神經發育。本研究主要的目的是建立在早產兒神經學發展的動物模型，以探討產後單一劑量Dex對海馬迴結構神經元密度和神經膠細胞組成在發育上的影響。
實驗方法與材料：實驗動物為大白鼠(Wistar Albino rat)之自然出生幼鼠，懷孕期約21+1天，幼鼠出生後的24小時內定義為產後第一天(P1)。在P1時候，幼鼠隨機經腹腔給予單一劑量Dex (Oradexon, 4mg/mL, Organon, Netherlands)，或是等量生理食鹽水。所有幼鼠分別於產後P7（足月時期）, P21（離乳時期）, P35（青春期前期）時犧牲。全腦分離後固定於蠟包埋之中，冠狀切面系列切片(7μm厚度)，每六片進行Nissl 染色，在Leitz Aristroplan顯微鏡40x10倍率之下觀察，選取200μm的柵狀範圍，當切片為圖譜上Bregma前第31-33切片時，在海馬迴本體中的Ammon’s horn (CA1)和齒狀迴的上錐狀葉(suprapyramidal blade of dentate gyrus, S-DG) 做神經元細胞密度的定量計數。神經元密度根據切面為基礎之方法重複計數三次，求取平均值，資料分析利用SPSS統計軟體進行非母數的Mann-Whitney test或是Kruskal-Wallis test.
實驗結果分為兩部份：首先，為了探討產後單一Dex劑量高低的影響，實驗組幼鼠分別接受高劑量0.5 mg/kg (Dex-DH)和低劑量0.2mg/kg (Dex-DL)的治療，共有五胎46隻幼鼠參與實驗，控制組(NS)的平均體重在P1為6.79+0.39 g，在Ｐ7為15.21+0.49 g，在Ｐ21為57.08+2.18 g，在Ｐ35為144.18+17.40 g。平均腦重在P1為0.2991+0.0306 g, 在Ｐ7為0.6632+0.0462 g , 在Ｐ21為1.4444+ 0.0383 g , 在Ｐ35為1.6360+0.1445 g 。在實驗組Dex-DH和Dex-DL顯著減少P7-21-35平均體重、顯著減少P7-21平均腦重(P<0.03)，且重量減少程度和劑量較高有統計上意義，但是P35腦重在實驗組和對照組間無顯著差異。型態學上，在實驗組的S-DG和CA1都有出現核濃染細胞和變寬的神經元細胞間隙。定量來說，正常齒狀回的神經元密度在P1時83.0+5.3 per 10,000μm2，P14-P21日齡時會達到最高密度(P7, 107.9+3.8 per10,000μm2; P21, 114.7+3.0 per 10,000μm2)，之後維持穩定(P35, 127.3+6.0 per10,000μm2)；實驗組不論是在S-DG和CA1的海馬迴區域，P7-21-35都有顯著較低的神經元密度(p<0.02)，而且神經元密度減少的程度和劑量較高有統計上的差異（p＜0.05）。
接著，我們進一步探討在更密集的時間間隔進行免疫組織化學染色以探討Dex對神經膠細胞的組成變化。自然出生的幼鼠在Ｐ1接受單一劑量Dex 0.5 mg/kg或等量生理食鹽水注射，分別在出生後P3（早產時期）, P7（足月時期）, P14（近離乳時期）, P21（離乳時期）, P28（離乳後時期）, P35（青春期前期）犧牲，全腦分離做冷凍保護後進行冠狀切面冷凍切片(25μm 厚度)，在懸浮溶液中利用單株抗體進行下列免疫組織染色(1) nestin標定神經表皮幹細胞(2) vimentin標定輻射狀膠細胞(3)glial fibrillary acidic protein (GFAP)標定星狀細胞(4)MRC-OX42標定小神經膠細胞(5)Nissl染色觀察型態。每隻動物在中低倍率下觀察至少五片切片，根據染色分為下面四個等級：背景染色(•)、輕度反應(+)、中度反應(++)、和強烈反應(+++)。我們發現控制組的海馬迴結構P3-7時期，nestin、vimentin有強烈的反應，P14之後時期則反應訊號微弱。在GFAP, MRC-OX42一開始P3-7反應微弱，P14之後反應變的中度到強度反應。在實驗組，nestin反應訊號在P3-7受到抑制、P14-21-28-35只有背景染色；vimentin的反應在P3明顯被抑制，P7之後只有背景染色；在GFAP星狀細胞反應延遲到P28-35才有中度到強度反應；MRC-OX42的分支狀小神經膠細胞反應提早到P7明顯出現。
由上述形態學的觀察發現，Dex產後暴露會減少海馬迴結構神經元細胞的密度，而且減少神經表皮幹細胞、輻射狀膠細胞等神經前驅細胞的表現，延遲正常星狀細胞的表現，提早分支狀小神經膠細胞的表現。儘管只有單一劑量，Dex對神經發育的影響可以從早產時期、足月時期、離乳時期持續到青春期前期。
程式化效應定義為，在發展過程中，某種因子因為作用在發育過程中某一階段，卻能造成結構上的永久改變，甚至此改變會持續一生的影響。類固醇荷爾蒙對神經發育可能有程式化效應，發育中腦部接受外源性的Dex對神經前驅細胞產生抑制性的程式化效應，影響到神經元細胞和神經膠細胞的密度與訊號，造成海馬迴結構的長期改變。程式化效應可以幫忙了解臨床觀察研究中Dex暴露的早產兒，即使在學齡兒童時期仍有持續不良的神經預後。
綜合臨床的觀察與動物的實驗，早產兒腦部產後Dex的曝露對神經發育有不良影響，而且此影響可以持續到學齡兒童（臨床）和青春期前期（動物實驗），這可能是一種程式化效應。

關鍵字：腦部發育、產後治療、海馬迴結構、Dexamethasone、程式化效應。

關鍵字

腦部發育 ； 產後治療 ； 海馬迴結構 ； 程式化效應 ； Dexamethasone

並列摘要

Dexamethasone (Dex), a synthetic glucocorticoid, is potent and long duration. It decreases early and late manifestations in the inflammatory process and induces cytochrome P450 dependent enzymes. Chronic lung disease (CLD) is an important cause of mortality and morbidity in preterm infants. Because inflammation plays an important role in the pathogenesis of CLD, corticosteroids, in particular Dex, have been widely used to prevent or treat CLD.
However, postnatal Dex therapy to prevent CLD was reported to affect somatic growth and neurodevelopmental outcomes followed up at childhood in several clinical studies. We recently reported a group of our preterm infants who had participated in a placebo-controlled, double-blind trial of Dex therapy begun within 12 hours after birth, were more likely to have delays in somatic growth, impaired neuromotor and cognitive dysfunction, and disability at school age. (Appendix 1)
Hippocampal formation contains high density of glucocorticoid receptors, is the major inhibitory input of hypothalamic-pituitary-adrenal axis, and has functions of recognition development, hypothalamic control, emotion expression, memory and learning. Our hypothesis is the exposure to postnatal Dex therapy in prematurity may affect the organization of hippocampal formation. The aim of this study was to establish an animal model of developing brain for manipulation, to investigate the effects of a single dose of postnatal Dex on the neuron density and organization of neuroglial cells of hippocampal formation in rat pups.
Time-dated Wistar albino pregnant rats were allowed to deliver naturally. Pups were delivered naturally on gestation 21+1 days. The first 24 hour after birth is defined as postnatal day 1 (P1). On P1, pups randomly received either single dose of Dex (Oradexon, 4mg/mL, Organon, Netherlands) or equivalent volume of normal saline intraperitoneally. All the pups were sacrificed on P7 (full-term), P21 (weaning) and P35 (prepubertal period), respectively. Whole brains were removed and embedded in paraffin, then were serially coronally sectioned (7 μm in thickness). Every sixth section was stained with cresyl violet for morphological observation (Nissl-stained). We counted the neuron density using a grid of 200μm x 50μm under 40x10 magnification by a Leitz Aristroplan microscope. Ammon’s horn (CA1) and suprapyramidal blade of middle dentate gyrus (S-DG) at Bregma Plate 31-33 were chosen for comparing the density of neurons between the control and Dex treated pups. The neuron density was counted in triplicate by the profile-based method. Data were analyzed by non-parametric Mann-Whitney test or Kruskal-Wallis test for comparing the differences between means or variance by SPSS statistical software.
The results consisted of two parts. In order to study the dose-dependent effect of single pharmacological Dex, every neonate rat in treated groups received Dex 0.5 mg/kg/dose (dose high, Dex-DH) or 0.2 mg/kg/dose (dose low, Dex-DL). 46 Wistar rat pups from 5 litters were analyzed with at least five pups in each group. Their mean body weight were 6.79+0.39 g at P1, 15.21+0.49 g at P7, 57.08+2.18 g at P21, 144.18+17.40 g at P35 in control (NS) group. Their mean brain weight were 0.2991+0.0306 g at P1, 0.6632+0.0462 g at P7, 1.4444+ 0.0383 g at P21, and 1.6360+0.1445 g at P35 in NS group. Both Dex-DL and Dex-DH groups, compared to NS group, had significantly lesser body weight at P7-21-35, and lower brain weight at P7-21, and had dose-dependent effect (p <0.03). But the brain weight at P35 between treated groups and control groups was not significantly different. Morphologically, there were increased pyknotic cells, widening of intercellular space in S-DG and CA1 in Dex-treated groups. Quantitatively, normal neuron density of suprapyramidal-blade dentate gyrus was 83.0+5.3 per 10,000μm2 at P1, peak density at P14-P21(P7, 107.9+3.8 per10,000μm2; P21, 114.7+3.0 per 10,000μm2)，then maintained stable density at P53(127.3+6.0 per10,000μm2). Dex-DL, and Dex-DH had less number of neuron density in S-DG and CA1 at P7-21-35 than control (p <0.02). In addition, the degree of neuronal loss was significantly more prominent in pups received higher dose of Dex (p <0.05).
Then, we conducted the second study of immunohistochemical (IHC) staining with frequent intervals to investigate the change of neuroglial cells in the hippocampal formation. After natural delivery, pups randomly received either single dose of Dex (0.5 mg/kg) or equivalent volume of normal saline intraperitoneally. All the pups were sacrificed on P3 (preterm), P7 (full-term), P14 (near-weaning), P21 (weaning), P28 (post-weaning) and P35 (prepubertal period), respectively. Brains were removed, post-fixed and cryoprotected in 30% sucrose/PBS. Segments were freeze-mounted in embedding medium and cryostat sectioning (25 μm). IHC were performed on free floating sections with following monoclonal antibodies. (1) nestin for neuroepithelial stem cell, (2) vimentin for radial-like glial cells, (3) glial fibrillary acidic protein (GFAP) for mature astrocyte, (4) MRC-OX 42 for microglia, (5) Nissl stain for morphological observation. Following staining, five or more sections from each animals were surveyed under low and medium magnification to arrive at scoring based on following scale: baseline (•), mild response (+), moderate response (++), intense response (+++). Criteria for each score were described. In Dex groups, we found the nestin-immunoreactive (IR) signals was inhibited on P3-7, then becoming background staining on P14-21-28-35; vimentin-IR signals was inhibited on P3, then becoming background staining after P7; GFAP-IR astrocyte did not present until P28-P35; MRC-OX 42-IR ramified microglia prematurely presented since P7.
By morphological observation, Dex exposure decreased neuron density, inhibited presentation of neuroprecusor cells, delayed mature astrocyte, but prematurely presented ramified microglia. Although only single-dose of pharmacological dose, the effect of Dex on neurological development persisted from preterm—term—weaning—prepubertal periods.
Programming effect is defined as the ability of a factor, acting during a defined developmental stage, to exert organizational changes that persist through life. Glucocorticoids probably have programming effects. Exogenous Dex exposure during developing brain may exert inhibiting programming effect on somatic growth, inhibiting neural precursors, decreased presentations of neurons and neuroglials, causing long-term organization change of hippocampal formation. It is implicated with clinical observation that Dex-exposed preterm infants suffered from persisting deleterious outcome at school age.
In brief, both clinical observations and our animal study showed postnatal Dex exposure in prematurity have deleterious effects on neurological development, and the effects persisted to school-aged (in clinical study) and prepubertal periods (in animal study). This is probably a programming effect.

Key words: brain development, postnatal therapy, hippocampal formation, dexamethasone, programming effect.

並列關鍵字

brain development ； postnatal therapy ； dexamethasone ； hippocampal formation ； programming effect

參考文獻

Anonymous. Effect of corticosteroids for fetal maturation on perinatal outcomes. NIH Consensus Development Panel on the Effect of Corticosteroids for Fetal Maturation on Perinatal Outcomes. JAMA. 1995;273(5):413-8.

Avery GB, Fletcher AB, Kaplan M, Brudno DS. Controlled trial of dexamethasone in respirator-dependent infants with bronchopulmonary dysplasia. Pediatrics. 1985;75(1):106-11.

Azmitia EC, Jr., McEwen BS. Adrenalcortical influence on rat brain tryptophan hydroxylase activity. Brain Research. 1974;78(2):291-302.

Ballard PL, Ballard RA. Scientific basis and therapeutic regimens for use of antenatal glucocorticoids. American Journal of Obstetrics & Gynecology. 1995;173(1):254-62.

Barker DJ. In utero programming of chronic disease. Clinical Science. 1998;95(2):115-28.