|摘要: ||虎杖(Polygonum cuspidatum)，為一種傳統上常用的中藥材，具有有祛風利濕、止咳化痰、鎮痛解毒、活血祛瘀之功效，可用來治療關節痛、慢性支氣管炎、高血壓、高膽固醇血症、外傷、炎症等等疾病。根據前人研究的發現，虎杖的根部抽出物可以分離出白藜蘆醇、白藜蘆醇苷及大黃素這三大主成分，由於這些具有生理活性的多酚類化合物存在，致使虎杖能擁有良好的藥理作用。白藜蘆醇及白藜蘆醇苷已被發現到具有抗氧化、抗發炎、抗癌及預防心血管疾病的功效，大黃素也已被發現到其具有像是抗菌、抗發炎及抗腫瘤增生等等功效。鑑於中藥材虎杖具有許多良好的生理活性，本論文將以萃取及生物催化來切入，進行三部份之研究。|
第一部份：超音波輔助萃取虎杖中的白藜蘆醇、白藜蘆醇苷及大黃素。為了萃取出植物體內的活性成分，希望藉由超音波作用時對液體所產生的空穴效應來提高萃取的效率，此部份的研究將藉由質傳效果的分析，來比較超音波輔助萃取和傳統搖瓶萃取法，並證實超音波輔助萃取之效益，本部份研究利用反應曲面法配合Box-Behnken三階層三變數之實驗設計，評估溫度、乙醇濃度及超音波功率對虎杖中的白藜蘆醇、白藜蘆醇苷及大黃素萃取產率之影響，找出最適化之萃取條件 (在最適化條件之下，白藜蘆醇苷、白藜蘆醇及大黃素之萃取產率分別為10.77、3.82及11.72 mg/g)。
第二部份：酵素催化白藜蘆醇苷轉化為白藜蘆醇之最適化探討。在自然界中，白藜蘆醇通常以其糖苷衍生物—白藜蘆醇苷的形式存在於植物體內，所以白藜蘆醇苷的含量往往都比白藜蘆醇高出許多，但白藜蘆醇苷的生物可利用性卻不及白藜蘆醇，因此希望能開發出能將白藜蘆醇苷轉化成白藜蘆醇的技術，用以改善其生物可利用性。此部份的研究是利用生物催化的方式，以商業化複合酵素ViscozymeR L之β-葡萄糖苷水解酶活性，來催化白藜蘆醇苷轉化成白藜蘆醇，並利用反應曲面法結合中心混成實驗設計來評估溫度、時間、酵素用量以及pH值對轉化率之影響，獲得最適化的催化條件(在溫度50℃、反應時間4.75 h、酵素用量2.5 FBG及pH值4.3之條件下進行反應，白藜蘆醇之轉化率可達100%)。此外，鑑於虎杖中具有高含量之白藜蘆醇苷，因此也以此最適化之催化條件，應用於虎杖水萃液中，以生物催化的方式成功的將內含之白藜蘆醇苷轉化為白藜蘆醇。
第三部份：酵素輔助萃取虎杖中的白藜蘆醇。相較於傳統的萃取方法，酵素輔助萃取法是利用糖苷水解酶的作用，在萃取的過程中協助破壞植物體的細胞壁，以利有效成分的釋出，降低有機溶劑的使用，節省能源消耗及萃取時間，此部份的研究是利用商業化複合糖苷水解酵素來進行輔助萃取，希望有效萃取出虎杖中的白藜蘆醇並同時提高白藜蘆醇的產量，比較酵素的使用與否及種類對白藜蘆醇萃取產率的差異，以五階層四變數之中心混成實驗設計，探討溫度、時間、酵素用量以及pH值對萃取產率之影響，並以此實驗設計結合反應曲面法及類神經網路，評估兩者模組之適切性，再進而獲取最適化之萃取條件及產率(在溫度55℃、反應時間5 h、酵素用量395 PGNU及pH值5.4之條件下進行萃取，白藜蘆醇之萃取產率可達11.88 mg/g)。
Polygonum cuspidatum is a traditional Chinese herbal medicine. The roots of P. cuspidatum have been used in treatments for arthralgia, chronic bronchitis, hypertension, hypercholesterolemia, trauma and inflammation. The known compounds of resveratrol, piceid, and emodin are mainly constituent in the roots extracts of P. cuspidatum. The pharmacological activities of P. cuspidatum are mostly associated with these compounds. Resveratrol and piceid have been shown to carry various biological activities such as antioxidant capacity, anti-inflammatory, anti-cancer, and cardioprotective activities, whereas emodin has been shown to possess pharmaceutical properties such as anti-inflammatory, antibacterial, and anti-tumor activities. In view of P. cuspidatum with variety of biological activities, the studies of extraction bioactivity compounds and biotransformation process were focus in this research. It could be divided into three parts described as below.
Part 1: Ultrasonic-assisted extraction of piceid, resveratrol, and emodin from Polygonum cuspidatum was performed. In order to extract the bioactive compounds from plants, ultrasonic-assisted extraction technique was employed in this part. The ultrasound could create cavitations in the liquid solution to increase the efficiency of extraction. Besides, the mass transfer effects were evaluated for comparing the extraction efficiency of ultrasonic-assisted extraction with shaking extraction. The effect of temperature, ethanol concentration and ultrasound power on extraction yield of piceid, resveratrol, and emodin were evaluated by response surface methodology using a 3-variables-3-levels box-behnken design. Thus, the extraction conditions could be optimized for each compound.
Part 2: Enzymatic transformation of piceid to form aglycone resveratrol using β-glucosidase was studied. Generally, resveratrol is generally present in its glycoside form in nature plants. Piceid is a major glycoside form derivatives of resveratrol. The content of piceid is usually higher than its aglycone, resveratrol. However, the bioavailability of piceid is lower than resveratrol. Therefore, it is important to develop a technique to transform piceid into resveratrol by deglycosylation for improving its bioavailability. In order to achieve this purpose, a commercial enzyme ViscozymeR L was employed for transforming piceid to resveratrol in this part. The main object of this part was to optimize the enzymatic transformation conditions. The effect of temperature, time, enzyme amount and pH was examined by using response surface methodology with a central composite design. Furthermore, the optimal condition was applied to transform piceid in the water extracts of Polygonum cuspidatum roots to resveratrol.
Part 3: Optimization of enzymatic-assisted extraction of resveratrol from Polygonum cuspidatum by using response surface methodology(RSM) and artificial neural network (ANN) was studied. The principle of enzymatic-assisted extraction was utilized glycosidase to degrade or break down the plant cell wall for enhancement of the target component released from plant sources in extraction process. Comparing with traditional extraction method, enzymatic-assisted extraction not only decreased the solvent usage and energy consumption, but also increased the extraction yield of bioactive compounds. A commercial glycosidase complex, ViscozymeR L, was employed for enzymatic-assisted extraction process in order to increase the extraction yield of resveratrol. In addition, the effect on extraction yield of resveratrol with or without enzyme treatment was compared. The effect of temperature, time, enzyme amount and pH was examined by using 4-variables-5-levels central composite experiment design which was combined with response surface methodology or artificial neural network. The fitness of response surface methodology and artificial neural network was also compared. Finally, the optimal extraction conditions could be attained by using suitable prediction model.