基于高效液相色譜指紋圖譜的瑪咖及其制品真實(shí)性識(shí)別

劉興勇，邵金良，陳興連，王 麗，黎其萬(wàn)，劉宏程

（云南省農(nóng)業(yè)科學(xué)院質(zhì)量標(biāo)準(zhǔn)與檢測(cè)技術(shù)研究所，昆明 650223；農(nóng)業(yè)部農(nóng)產(chǎn)品質(zhì)量監(jiān)督檢驗(yàn)測(cè)試中心，昆明 650223）

基于高效液相色譜指紋圖譜的瑪咖及其制品真實(shí)性識(shí)別

劉興勇，邵金良，陳興連，王 麗，黎其萬(wàn)，劉宏程※

（云南省農(nóng)業(yè)科學(xué)院質(zhì)量標(biāo)準(zhǔn)與檢測(cè)技術(shù)研究所，昆明 650223；農(nóng)業(yè)部農(nóng)產(chǎn)品質(zhì)量監(jiān)督檢驗(yàn)測(cè)試中心，昆明 650223）

瑪咖及其制品的真實(shí)性是保障品質(zhì)的關(guān)鍵。為了研究瑪咖及其制品的差異，對(duì)瑪咖制品進(jìn)行真實(shí)性鑒別，分別以不同產(chǎn)地、色型的瑪咖及瑪咖制品為供試材料，對(duì)其進(jìn)行高效液相色譜（high performance liquid chromatography，HPLC）指紋圖譜不同識(shí)別方法的研究。采用Waters Symmetry ShieldTMC18（250 mm×4.6 mm，5 μm）色譜柱，乙腈-水為流動(dòng)相梯度洗脫，檢測(cè)波長(zhǎng)210 nm，體積流量0.80 mL/min，各成分得到較好分離。經(jīng)方法學(xué)驗(yàn)證，方法具有較好的精密度、重復(fù)性和穩(wěn)定性。對(duì)19個(gè)瑪咖樣品和6個(gè)瑪咖制品HPLC指紋圖譜進(jìn)行分析，確定了15個(gè)色譜峰為瑪咖和其制品的特征指紋峰，建立了瑪咖及其制品的指紋圖譜。以數(shù)字化指紋圖譜為基礎(chǔ)，分別進(jìn)行主成分分析、判別分析和相似度分析。結(jié)果表明，3種方法均能使瑪咖與瑪咖制品得到較為一致的模式識(shí)別結(jié)果。主成分二維平面圖和判別分析能夠區(qū)分瑪咖和瑪咖制品，具有簡(jiǎn)便、直觀的特點(diǎn)，瑪咖與瑪咖制品相似度分析差異顯著（P＜0.05），分別為0.916和0.668，不同來(lái)源和色型的瑪咖、瑪咖制品間相似度均無(wú)顯著差異（P＞0.05）。結(jié)果顯示瑪咖制品具有瑪咖的特征峰，但含量存在差異。3種方法均能準(zhǔn)確地體現(xiàn)指紋圖譜的一致性和特征性，為瑪咖和瑪咖制品的區(qū)分及瑪咖制品真實(shí)性保障提供了參考。利用HPLC指紋圖譜可對(duì)瑪咖及其制品的真實(shí)性進(jìn)行鑒別。

加工；模式識(shí)別；瑪咖；高效液相色譜；指紋圖譜

劉興勇，邵金良，陳興連，王 麗，黎其萬(wàn)，劉宏程.基于高效液相色譜指紋圖譜的瑪咖及其制品真實(shí)性識(shí)別 [J].農(nóng)業(yè)工程學(xué)報(bào)，2016，32（6）：302-307.doi：10.11975/j.issn.1002-6819.2016.06.042 http://www.tcsae.org

Liu Xingyong,Shao Jinliang,Chen Xinglian,Wang Li,Li Qiwan,Liu Hongcheng.Authenticity identification of Maca and its product based on high performance liquid chromatography fingerprint[J].Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE）,2016,32(6):302-307.(in Chinese with English abstract) doi：10.11975/j.issn.1002-6819.2016.06.042 http://www.tcsae.org

0 引言

瑪咖（Maca）Lepidium meyenii Walp.為十字花科植物，原產(chǎn)于海拔4 000～4 500 m的秘魯安第斯山脈地區(qū)[1]，營(yíng)養(yǎng)成分豐富，是一種食藥兼用植物，其中的酰胺[2]、膳食纖維[3]、芥子油苷和多酚[4]等功效成分增加了瑪咖的附加值，成為衡量其質(zhì)量的指標(biāo)。經(jīng)研究表明瑪咖根莖具有改善記憶、增強(qiáng)免疫、增強(qiáng)精子活力、抗疲勞、抗氧化等多種功效[5-8]，中國(guó)于2011年經(jīng)衛(wèi)生部第13號(hào)公告批準(zhǔn)瑪咖為新資源食品，并大規(guī)模引種栽培成功。新鮮瑪咖經(jīng)過(guò)一定加工工藝制成干片、粉末和藥劑等商品化產(chǎn)品，但由于產(chǎn)品失去瑪咖原有的感官特性及加工過(guò)程中添加其它假冒成分，從外觀上極難鑒別其真?zhèn)魏痛_保質(zhì)量。近年來(lái)中國(guó)瑪咖產(chǎn)業(yè)的快速發(fā)展導(dǎo)致缺乏質(zhì)量評(píng)價(jià)和真?zhèn)舞b別的方法和依據(jù)，市場(chǎng)較混亂。因此，如何保障瑪咖及其制品的品質(zhì)及真實(shí)性成為發(fā)展瑪咖產(chǎn)業(yè)亟待解決的問(wèn)題。

對(duì)瑪咖的真?zhèn)舞b別技術(shù)正在進(jìn)行探索性研究中。Jim[9-10]、Chen[11]等分別采用GC/MS（氣相色譜-質(zhì)譜聯(lián)用儀，gas chromatography-mass spectrometer）、近紅外光譜技術(shù)從揮發(fā)油的獨(dú)特圖譜特征和分子生物學(xué)技術(shù)對(duì)瑪咖和易混淆品進(jìn)行了鑒別和質(zhì)量評(píng)價(jià)，該技術(shù)適用于瑪咖原材料的鑒別。揮發(fā)性成分的含量相對(duì)較低，影響因素多，檢測(cè)分析對(duì)儀器要求較高。市場(chǎng)中的瑪咖制品往往是瑪咖粉碎后添加其它易混品制成，同樣含有瑪咖的獨(dú)特成分，如瑪咖酰胺、芥子油苷等提取物含量被作為衡量瑪咖及其產(chǎn)品質(zhì)量的主要標(biāo)準(zhǔn)之一[12]。研究表明，游離脂肪酸和芐胺是生成瑪咖酰胺的前體，瑪咖干燥過(guò)程中組織軟化破壞，釋放游離脂肪酸，芥子油苷內(nèi)源性代謝水解生成異硫氰酸酯的同時(shí)產(chǎn)生大量芐胺[2]?，斂е薪孀佑蛙蘸渴艿綐悠飞?、狀態(tài)、干燥方式等影響[13]，從而導(dǎo)致瑪咖酰胺含量產(chǎn)生差異，且芥子油苷廣泛存在于瑪咖莖、葉和十字花科植物中[14]。因此，不能單獨(dú)從瑪咖酰胺和芥子油苷等含量來(lái)判別瑪咖及其制品的真實(shí)性。針對(duì)瑪咖及其制品真?zhèn)舞b別的報(bào)道較鮮見(jiàn)，需要建立廣泛適用的方法?，斂Ъ捌渲破肪鶠槎嘟M分復(fù)雜體系，因此鑒別其真實(shí)性應(yīng)采用與之相適應(yīng)的，且能提供豐富鑒別信息的檢測(cè)方法。通過(guò)建立高效液相色譜指紋圖譜能較為全面地反映瑪咖及其制品中化學(xué)成分的種類與數(shù)量，進(jìn)而對(duì)質(zhì)量進(jìn)行整體描述和評(píng)價(jià)。相關(guān)學(xué)者采用高效液相色譜法對(duì)瑪咖中的酰胺類、瑪咖烯和芥子油苷等組分進(jìn)行了研究[15-16]，為從HPLC（高效液相色譜，high performance liquid chromatography）指紋圖譜來(lái)鑒別瑪咖及其制品的真?zhèn)魏捅Ｕ腺|(zhì)量提供了參考，并且高效液相色譜技術(shù)方便快捷，已廣泛應(yīng)用在茶葉[17-18]、中藥[19]等鑒別和評(píng)價(jià)。

本研究以瑪咖及其制品為試驗(yàn)對(duì)象，旨在建立瑪咖及其商品化產(chǎn)品HPLC指紋圖譜，并通主成分分析、判別分析和相似度評(píng)價(jià)等識(shí)別方法和模式，對(duì)瑪咖及其制品指紋圖譜進(jìn)行分析，為瑪咖及其制品的區(qū)分及瑪咖制品真?zhèn)舞b別提供參考。

1 材料與方法

1.1 儀器、試劑與材料

Waters Alliance 2695高效液相色譜儀（美國(guó)Waters公司），包括四元泵溶劑淋洗系統(tǒng)、自動(dòng)進(jìn)樣系統(tǒng)，2487紫外檢測(cè)器，Empower色譜工作站；Laborata 4000-efficient旋轉(zhuǎn)蒸發(fā)儀（德國(guó)Heidolph公司）；AE100型電子分析天平（梅特勒-托利多儀器有限公司）；KQ500-E型超聲清洗器（江蘇昆山市超聲儀器有限公司）。乙腈（德國(guó)Merck公司）色譜純；哇哈哈礦泉水；石油醚（國(guó)藥集團(tuán)）分析純?，斂Мa(chǎn)地為云南麗江、香格里拉、會(huì)澤、尋甸、四川雅安、西藏曲水、林周縣，并經(jīng)鑒定確認(rèn)為瑪咖的地下根莖部分，瑪咖制品為瑪咖干片、片劑，購(gòu)自市場(chǎng)，具體來(lái)源見(jiàn)表1。

表1 瑪咖來(lái)源信息Table 1 Information about sources of Maca

1.2 試驗(yàn)方法

1.2.1 液相色譜條件

色譜柱Waters Symmetry ShieldTM（C18，5 μm，4.6×250 mm）；進(jìn)樣量10.0 μL；柱溫30.0℃；檢測(cè)波長(zhǎng)210 nm；流速為0.80 mL/min；流動(dòng)相A為水，流動(dòng)相B為乙腈；瑪咖石油醚提取成分復(fù)雜，經(jīng)條件優(yōu)化處理，將梯度洗脫程序設(shè)計(jì)為：0～10 min，80%A→10～35 min，80%～60%A→35.0～40.0 min，60%～40%A→40～45 min，40%～35%A→45～50 min，35%～20%A→50～55 min，20%～60%A→55～65 min，60%～80%A。

1.2.2 供試液的制備

曬干樣品和制品用粉碎機(jī)粉碎過(guò)60目篩，依據(jù)瑪咖熱風(fēng)干燥特性[20]于電熱鼓風(fēng)干燥箱中70℃干燥至恒質(zhì)量，自封袋密封保存于干燥器皿中。供試液制備參照文獻(xiàn)[15]，精確稱取烘干的粉末約1.000 g，置于100 mL錐形瓶中，加入石油醚10 mL，200 MHz超聲30 min，放置至室溫后過(guò)濾，旋轉(zhuǎn)蒸發(fā)儀濃縮至干，加入5 mL乙腈溶解，經(jīng)0.45 μm微孔濾膜過(guò)濾備用。

1.2.3 數(shù)據(jù)處理

將樣品按給定的色譜條件進(jìn)行操作，記錄色譜圖如圖1，橫坐標(biāo)表示色譜峰的出峰時(shí)間，縱坐標(biāo)表示色譜峰的響應(yīng)值大小，建立指紋圖譜，指紋圖譜的評(píng)價(jià)采用相似度進(jìn)行。相似度反映了不同樣品間在某種恒定分析情況下特定成分在含量及組成上的相似程度，常用來(lái)對(duì)中藥進(jìn)行質(zhì)量及真?zhèn)蔚蔫b別。相似度計(jì)算常用夾角余旋法，夾角余弦是將指紋圖譜數(shù)據(jù)視為多維空間向量，使指紋圖譜的相似性問(wèn)題轉(zhuǎn)化為多維空間中向量間的相似問(wèn)題,計(jì)算按公式（1）進(jìn)行。

式中xi′為樣品圖譜中各共有峰峰面積；yi′為對(duì)應(yīng)的標(biāo)準(zhǔn)圖譜中每個(gè)組分峰面積。以所有樣品中都出現(xiàn)的峰為共有峰，采用均值法產(chǎn)生標(biāo)準(zhǔn)圖譜R，以樣品中每個(gè)共有峰峰面積平均值為對(duì)應(yīng)標(biāo)準(zhǔn)圖譜峰面積。采用“中藥色譜指紋圖譜相似度評(píng)價(jià)系統(tǒng)A版”軟件繪制圖譜疊加圖。用SPSS17.0軟件進(jìn)行主成分分析和判別分析來(lái)評(píng)價(jià)指紋圖譜對(duì)瑪咖及其制品質(zhì)量評(píng)價(jià)及真?zhèn)舞b別的可靠性。

2 結(jié)果與分析

2.1 方法學(xué)試驗(yàn)

通過(guò)對(duì)同一批樣品進(jìn)行精密度試驗(yàn)、穩(wěn)定性試驗(yàn)、重復(fù)性試驗(yàn)，以色譜峰9為參照峰，各主要色譜峰相對(duì)保留時(shí)間RSD（relative standard deviation）值均分別小于3.5%，相對(duì)峰面積RSD值小于5%，表明儀器精密度良好，重復(fù)性好，供試液在24 h內(nèi)穩(wěn)定。

2.2 指紋圖譜的建立

將樣品供試液按上述HPLC指紋圖譜方法測(cè)定，將19份瑪咖樣品粉末及6份瑪咖制品按給定的方法條件進(jìn)行操作，記錄HPLC色譜圖如圖1。圖1為共有峰圖譜，橫坐標(biāo)表示色譜峰出峰時(shí)間，縱坐標(biāo)表示色譜峰響應(yīng)值大小，共標(biāo)定15個(gè)色譜峰作為HPLC指紋圖譜的共有峰，15個(gè)共有峰的面積占總峰面積在92.5%～97.8%之間，其中9號(hào)峰分離完全，出峰時(shí)間適中，峰面積較大，在各樣品中均存在，設(shè)為參照峰。

圖1 瑪咖共有峰HPLC圖譜Fig.1 HPLC of common peak of Maca

2.3 瑪咖及其制品指紋圖譜的模式識(shí)別

2.3.1 主成分分析

主成分分析（principal component analysis，PCA）是對(duì)多變量數(shù)據(jù)進(jìn)行降維處理的一種方法，可用較少的變量去解釋原來(lái)樣品中的大部分變量，選出比原始變量個(gè)數(shù)少又能解釋大部分樣品變量的幾個(gè)新變量，并用以解釋樣品的綜合性和差異性。為驗(yàn)證所建指紋圖譜對(duì)瑪咖與瑪咖制品區(qū)別的可靠性，以HPLG指紋圖譜中標(biāo)定的15個(gè)共有峰峰面積原始數(shù)據(jù)為變量，用SPSS 17.0統(tǒng)計(jì)分析軟件對(duì)數(shù)據(jù)進(jìn)行主成分分析，依據(jù)主成分個(gè)數(shù)提取原則提取特征根大于1的主成分，得到成分載荷矩陣結(jié)果見(jiàn)表2。成分載荷矩陣中每一列數(shù)值表示相關(guān)變量與提取主成分間的相關(guān)系數(shù)，其絕對(duì)值越大，則主成分對(duì)該變量的代表性也越大。15個(gè)共有峰中共提取5個(gè)主成分，累計(jì)貢獻(xiàn)率為100%。第1主成分特征值為6.384，方差貢獻(xiàn)率為42.558%，包含了共有峰1、2、3、6、7、9、10、11和13的信息；第2主成分特征值為3.443，方差貢獻(xiàn)率為22.956%，包含峰5、8和12的信息；第3主成分特征值為2.523，方差貢獻(xiàn)率為16.818%，代表峰15的信息；第4主成分特征值為1.568，方差貢獻(xiàn)率為10.450%，代表峰14的信息；第5主成分特征值為1.083，方差貢獻(xiàn)率為7.218%，代表峰4的信息。從累計(jì)貢獻(xiàn)率來(lái)看，取前4個(gè)主成分時(shí)，累計(jì)貢獻(xiàn)率為92.78%。根據(jù)主成分分析原理，取前4個(gè)為主要成分足可以說(shuō)明各指標(biāo)的信息。以表2第1、2主成分載荷建立PCA因子分布平面圖，見(jiàn)圖2，除制品中的S2樣品外，瑪咖和瑪咖制品可以區(qū)分開(kāi)，說(shuō)明二者差異主要集中在前2個(gè)主成分中。

表2 成分載荷矩陣Table 2 Component loading matrix

圖2 供試樣品指紋圖譜主成分分析Fig.2 Principal component analysis of fingerprints obtained from samples

2.3.2 瑪咖及其制品判別分析

線性判別分析（linear discriminant analysis，LDA）是以部分樣品數(shù)據(jù)信息建立判別模式，再用判別模式對(duì)另一部分樣品進(jìn)行驗(yàn)證，從而達(dá)到區(qū)分鑒別目的的一種數(shù)據(jù)分析方法，常與主成分分析相互補(bǔ)充應(yīng)用。以指紋圖譜中標(biāo)定的共有峰原始峰面積為變量，用SPSS17.0軟件進(jìn)行判別分析，將分類樣品設(shè)為瑪咖和制品兩類，共有峰峰面積為自變量，采用留一法交叉驗(yàn)證，將每一個(gè)樣本作為測(cè)試樣本，其它樣本作為訓(xùn)練樣本來(lái)衡量分類結(jié)果。輸出預(yù)測(cè)分類結(jié)果見(jiàn)表3?，斂Ш同斂е破吩诔跏寂袆e中正確率均為100%，交叉驗(yàn)證中瑪咖有一個(gè)樣品（S7）誤判為瑪咖制品，正確率為94.7%，對(duì)瑪咖制品的判別正確率為100%，整體正確率為97.4%。判別分析結(jié)果與主成分分析結(jié)果高度一致，進(jìn)一步證明所建瑪咖和瑪咖制品指紋圖譜具有較好代表性。

表3 判別分析結(jié)果Table 3 Result of discriminant analysis

2.3.3 相似度分析

為進(jìn)一步驗(yàn)證主成分和判別分析的結(jié)果，對(duì)指紋圖譜進(jìn)行相似度分析。相似度可通過(guò)數(shù)字化形式直接反映出瑪咖與瑪咖制品的差異。相似度計(jì)算結(jié)果見(jiàn)表4。方差分析結(jié)果表明，瑪咖和瑪咖制品相似度差異較大（P< 0.05），瑪咖相似度平均值為0.916，除S7外都在0.80以上，瑪咖制品相似度平均值為0.668，除S2外相似度均在0.80以下。相似度分析結(jié)果在主成分分析和判別分析中均得以體現(xiàn)。不同瑪咖制品間相似度差異不顯著（P＞0.05），瑪咖干片平均相似度為0.737，片劑為0.598。不同產(chǎn)地瑪咖相似度差異不顯著（P＞0.05），會(huì)澤、麗江、曲水分別為0.882，0.928和0943。不同色型瑪咖相似度差異不顯著（P＞0.05），黃、紫、黑瑪咖分別為0.920、0.922和0.904。不同產(chǎn)地及色型的瑪咖相似度均無(wú)顯著差異，說(shuō)明所建指紋圖譜具有針對(duì)性和整體性。從圖3a指紋圖譜疊加圖可看出瑪咖制品共有色譜峰峰面積較小，說(shuō)明瑪咖在加工過(guò)程中可能損失了相關(guān)組分。規(guī)?；斂Ц稍锒嗖捎米匀涣栏桑瑴囟认鄬?duì)較低，細(xì)胞不易損傷，對(duì)瑪咖中成分保存較好[20]，而真空干燥由于細(xì)胞的破損，溫度的波動(dòng)及黑芥子酶對(duì)芥子油苷的水解作用使芥子油苷顯著降解[13，22]。從圖3b的不同色型瑪咖與瑪咖制品的對(duì)照?qǐng)D譜來(lái)看，不同色型瑪咖共有峰一致，而制品色譜圖中出現(xiàn)許多非共有峰，說(shuō)明加工中可能添加了其他成分。因此，通過(guò)指紋圖譜相似度分析可以用來(lái)鑒別瑪咖及其制品真實(shí)性。

表4 瑪咖指紋圖譜相似度Table 4 Similarity evaluation on Maca

圖3 瑪咖及其制品HPLC指紋圖譜Fig.3 HPLC fingerprint for Maca and products

通過(guò)瑪咖HPLC指紋圖譜中共有峰的存在可對(duì)瑪咖制品真實(shí)性進(jìn)行識(shí)別，但指紋圖譜無(wú)法反映制品中瑪咖的含量，且瑪咖制品中有效成分的含量影響因素眾多[12，20]，制品在沒(méi)有添加其他成分情況下，由于加工工藝等影響也可導(dǎo)致有效成分含量降低而存在差異。將來(lái)需要結(jié)合質(zhì)譜等手段進(jìn)一步研究瑪咖中添加外來(lái)成分有效成分含量變化，確定制品中瑪咖的比例等。

3 結(jié)論

瑪咖制品與瑪咖具有共同指紋峰，可以反映瑪咖制品的真實(shí)性。通過(guò)瑪咖和瑪咖制品的HPLC指紋圖譜，二者能夠較好區(qū)分，說(shuō)明二者在相關(guān)成分上具有差異。以主成分分析、判別分析和相似度分析三種模式進(jìn)行識(shí)別，結(jié)果顯示利用主成分分析和判別分析可以區(qū)分瑪咖和瑪咖制品。相似度分析表明瑪咖和瑪咖制品相似度差異較大（P＜0.05），不同來(lái)源、色型瑪咖之間、瑪咖制品之間指紋圖譜相似度無(wú)顯著差異（P＞0.05）。所建指紋圖譜具有較好的穩(wěn)定性和代表性，可用于瑪咖及其制品的真實(shí)性識(shí)別。

[1]Gonzales G F,Cordova A,Vega K,et al.Effect of Lepidium meyenii(Maca),a root with aphrodisiac and fertility-enhancing properties,on serum reproductive hormone levels in adult healthy men[J].Journal Endocrinology,2003,176(1):163-168.

[2]Eliana E,Antonella H,Kofer W,et al.Bioactive maca(Lepidium meyenii)alkamides are a resultoftraditionalAndean postharvest drying practices[J].Phytochemistry,2015,116(8): 138-148.

[3]Chen J J,Zhao Q S,Wang L W,et al.Physicochemical and functional properties of dietary fiber from maca(Lepidium meyenii Walp.)liquor residue[J].Carbohydrate Polymers,2015, 32(5):509-512.

將上述所得數(shù)據(jù)均帶入SPSS23.0統(tǒng)計(jì)學(xué)軟件中，將兩組患者的并發(fā)癥發(fā)生率作為計(jì)數(shù)資料，采用（%）表示，行χ2檢驗(yàn)；將患者的血糖水平、生活質(zhì)量評(píng)分作為計(jì)量資料采用（±s），行 t檢驗(yàn)；P＜0.05 為差異有統(tǒng)計(jì)學(xué)意義。

[4]David C,Rosana C,Omar B,et al.Optimized methodology for the simultaneous extraction of glucosinolates, phenolic compounds and antioxidant capacity from maca(Lepidium meyenii)[J].Industrial Crops and Products,2013,49:747-754.

[5]Rubio J,Caldas M,Davila S,et al.Effect of three different cultivars of Lepidium meyenii(Maca)on learning and depression in ovariectomized mice[J].BMC Complementary and Alternative Medicine,2006,6(1):23-29.

[6]JunLee K,Dabrowski K,Rinchard J,et al.Supplementation of maca(Lepidium meyenii)tuber meal in diets improves growth rate and survival of rainbow trout Oncorhynchus mykiss (Walbaum)alevins and juveniles[J].Aquaculture Research, 2004,35(3):215-223.

[7]Gonzales C,Rubio J,Gasco M,et al.Effect of short-term and long-term treatments with three ecotypes of Lepidium meyenii (MACA)on spermatogenesis in rats[J].Journal of Ethnopharmacology, 2006,103(3):448-454.

[8] 何釗，馮穎，徐瓏峰，等.云南種植瑪咖乙醇提取物的體外抗氧化活性分析[J].食品科學(xué)，2010，31(15)：39-43. He Zhao,Feng Ying,Xu Longfeng,et al.In vitro antioxidant activity of extract of maca(Lepidium meyenii walpers)cultivatedin Yunnan[J].Food Science,2010,31(15):39-43.(in Chinese with English abstract)

[9]Jin Wenwen,Xiong yue,Yu Longjiang.Identification and quality evaluation ofLepidium meyenii.(Maca)based on gas chromatographic analysis of its essential oils from roots[J]. Agricultural science&technology,2006,7(3):2-9.

[10]Jin W W,Zhang Y Z,Mei S,et al.Identification of Lepidium meyenii(Walp.)based on spectra and chromatographic characteristics of its principal functional ingredients[J].Journal of the Science of Food and Agriculture,2007,87(12):2251-2258.

[12]Melnikovova I,Havlik J,Cusimamani E F,et al.Macamides and fatty acids content comparison in maca cultivated plant under field conditions and greenhouse[J].Bol Latinoam Caribe Plant Med Aromat,2012,11(5):420-427.

[13]甘瑾，馮穎，張弘，等.三種色型瑪咖芥子油苷組分及含量分析[J].中國(guó)農(nóng)業(yè)科學(xué)，2012，45(7)：1365-1371. Gan jin,Feng ying,Zhang hong,et al.Analysis on composition and content of glucocinolate in three color types of maca (Lepidium meyenii)[J].Scientia Agricultura Sinica,2012,45(7): 1365-1371.(in Chinese with English abstract)

[14]Fahey J W,Zalemann A T,Talaley A T.The chemical diversity and distribution of glucosinolates and isothiocyanates among plants[J].Phytochemistry,2001,56(1):5-51.

[15]Ganzera M,Zhao J P,Muhammad I,et al.Chemical Profiling and Standardization of Lepidium meyenii(Maca)by Reversed Phase High Performance Liquid Chromatography[J].Chem.Pharm. Bull.2002,50(7):988-991.

[16]Li G,Ammermann U,Quirós C.Glucosinolate contents in maca (Lepidium peruvianum Chacón)seeds,sprouts,mature plants and several derived commercial products[J].Economic Botany, 2001,55(2):255-262.

[17]He X Y,Li J K,Zhao W,et al.Chemical fingerprint analysis for quality control and identification of Ziyang green tea by HPLC [J].Food Chemistry,2015,171(3):405-411.

[18]胡燕，齊桂年.南路邊茶高效液相色譜指紋圖譜的建立[J].現(xiàn)代食品科技，2013，29(9)：2283-2287. HU Yan,QI Guinian.Establishment of high performance liquid chromatographicfingerprint of southern route tea[J].Modern Food Science and Technology,2013,29(9):2283-2287.(in Chinese with English abstract）

[19]陳東東，周萍，白鋼鋼，等.基于HPLC中藥指紋圖譜技術(shù)延胡索藥材及其制劑的質(zhì)量控制探討[J].中國(guó)中藥雜志，2015，40(12)：2470-2473. Chen Dongdong,Zhou Ping,Bai Ganggang,et al.Discussion of HPLC fingerprint of traditional Chinese medicine of Corydalis yanhusuo and its preparation[J].China Journal of Chinese Materia Medica,2015,40(12):2470-2473.(in Chinese with English abstract)

[20]張弘，鄭華，于連松，等.瑪咖真空干燥特性及工藝參數(shù)優(yōu)化[J].農(nóng)業(yè)工程學(xué)報(bào)，2012，28(增刊1)：267-272. Zhang Hong,Zheng Hua,Yu Liansong,et al.Vacuum drying characteristics and technological parameters optimization of maca[J].Transactions of the Chinese Society of Agricultural Engineering(Transactions of the CSAE),2012,28(Supp.1):267-272.(in Chinese with English abstract)

[21]涂行浩，張弘，鄭華，等.貯藏溫度對(duì)瑪咖根塊采后品質(zhì)的影響[J].食品科學(xué)，2013，34（2）：282-287. Tu Xinghao,Zhang Hong,Zheng Hua,et al.Effect of storage temperature on postharvest of maca tuber[J].Food Science, 2013,34(2):282-287.(in Chinese with English abstract)

[22]Yábar E,Pedreschi R,Chirinos R,et al.Glucosinolate content and myrosinase activity evolution in three maca(Lepidium meyeniiWalp.)ecotypesduring preharvest,harvestand postharvest drying[J].Food Chemistry,2011,127(4):1576-1583.

Authenticity identification of Maca and its product based on high performance liquid chromatography fingerprint

Liu Xingyong,Shao Jinliang,Chen Xinglian,Wang Li,Li Qiwan,Liu Hongcheng※
(Institute of Quality Standards and Testing Technology,Yunnan Academy of Agricultural Sciences,Kunming 650223,China;Supervision＆Testing Center for Farm Products Quality,Ministry of Agriculture,Kunming 650223,China)

Maca has many health benefits and high price in China and the products may be adulterated,so authenticity is important to its quality.In order to identify the authenticity of maca products and analyze difference of maca and its products,chromatographic fingerprint peaks identification and difference analysis were used at Yunnan Academy of Agricultural Sciences(YAAS),Kunming city,Yunnan province,China,in 2015.Common maca products including tablets and dry film were purchased on Kunming markets and different ecotypes and regions maca collected from Lijiang,Tibet and Huize,which have big planting scale with representative.Fresh maca samples were by natural drying and smashed to powder less than 60 mesh,then dried to constant weight by drier at 70℃.About 1.000 g dried sample was ultrasonic extraction under 200 MHz with 10 mL petroleum ether for 30 min,then filtered with filter paper,rotary evaporator was usedto concentrate dry,added 5 mL acetonitrile to dissolve extract,and filtered by 0.45 μm filter membrane.Chromatographic peaks of maca were separated effectively by use A Waters Symmetry ShieldTMC18column with a detection wavelength of 210 nm and a gradient program with acetonitrile-water,and under 0.80 mL/min flow rate.According to the method validation,method has high precision,good repeatability and stability,and same batch maca samples chromatographic peaks relative retention time and relative peak area standard deviation relative values were less than 3.5%and 5%, respectively.19 batches of maca samples and 6 maca products fingerprint were analyzed,and the number 9 peak was set as reference peak to correct retention time because of separated complete,moderate appearance time,and larger peak area and all samples were existed.15 common characteristic peaks were determined of maca and its products basis retention time, their area accounted 92.5%-97.8%for total peak area,established the high performance liquid chromatography(HPLC) fingerprint of maca and its products,and digital fingerprint was also established based on relative retention time and corresponding peak area.Maca products common chromatographic peaks area were smaller compared with maca,and also contained some non-common peaks,this explained some components were lose in process and other components was added. Fingerprint for maca(Lepidium meyenii Walp.)and maca products was investigated based on three different recognition methods such as principal component analysis(PCA),discriminant analysis(LDA)and similarity analysis.PCA was performed based on digital fingerprint.PCA extracted characteristic root was greater than 1,then 5 principal components were extracted representing the information of sample,and cumulative contribution rate was 100%with component loading matrix.Two-dimensional scatter plot was drawn by use principal component 1 and 2 loading,so maca and its products can be distinguished by scatter plot.LDA divided samples into two categories,then chromatographic peaks area were as independent variable and one cross validation was used,and the classification result had 97.4%correct recognition rate for maca and its products.Maca samples could be distinguished obviously from maca product samples by using method of either two-dimensional map or LDA.Similarity were calculated by included angle,and similarity analysis indicated that maca and its products have difference(P≤0.05),and similarity value were 0.916 and 0.668,respectively,and not only no significant differences(P≥0.05)between different origins and ecotypes maca,but also between maca products.Maca from Huize,Lijiang,Tibet were 0.882,0.928 and 0943,respectively.Different ecotypes Maca of yellow,purple and blank were 0.920,0.922 and 0.904,respectively.Maca dry film and tablets were 0.737 and 0.598,respectively.The results showed that each of the three employed methods could reflect consistent pattern recognition results and characteristics of fingerprint. The results showed that maca products have characteristic components of maca,but their content had differences.This could provide an identification method and reference for guaranteeing the authenticity and quality of Maca products.

processing;pattern recognition;Maca;high performance liquid chromatography;fingerprint chromatography

10.11975/j.issn.1002-6819.2016.06.042

TS207.7

A

1002-6819（2016）-06-0302-06

2015-07-18

2016-01-11

云南省科技惠民專項(xiàng)（農(nóng)業(yè)）重點(diǎn)項(xiàng)目（2014RA054）

劉興勇（1985—），男，主要從事農(nóng)產(chǎn)品品質(zhì)研究。昆明 云南省農(nóng)業(yè)科學(xué)院質(zhì)量標(biāo)準(zhǔn)與檢測(cè)技術(shù)研究所，650223。

Emial：liuxingyong0993@163.com

※通信作者：劉宏程，博士，研究員，從事農(nóng)產(chǎn)品質(zhì)量與品質(zhì)分析檢測(cè)。昆明 云南省農(nóng)業(yè)科學(xué)院質(zhì)量標(biāo)準(zhǔn)與檢測(cè)技術(shù)研究所，650223。

Emial：liuorg@163.com