Analysis of Non-Volatile Chemical Constituents of Menthae Haplocalycis Herba by Ultra-High Performance Liquid Chromatography-High Resolution Mass Spectrometry

Menthae Haplocalycis herba, one kind of Chinese edible herbs, has been widely utilized for the clinical use in China for thousands of years. Over the last decades, studies on chemical constituents of Menthae Haplocalycis herba have been widely performed. However, less attention has been paid to non-volatile components which are also responsible for its medical efficacy than the volatile constituents. Therefore, a rapid and sensitive method was developed for the comprehensive identification of the non-volatile constituents in Menthae Haplocalycis herba using ultra-high performance liquid chromatography coupled with linear ion trap-Orbitrap mass spectrometry (UHPLC-LTQ-Orbitrap). Separation was performed with Acquity UPLC® BEH C18 column (2.1 mm × 100 mm, 1.7 μm) with 0.2% formic acid aqueous solution and acetonitrile as the mobile phase under gradient conditions. Based on the accurate mass measurement (<5 ppm), MS/MS fragmentation patterns and different chromatographic behaviors, a total of 64 compounds were unambiguously or tentatively characterized, including 30 flavonoids, 20 phenolic acids, 12 terpenoids and two phenylpropanoids. Finally, target isolation of three compounds named Acacetin, Rosmarinic acid and Clemastanin A (first isolated from Menthae Haplocalycis herba) were performed based on the obtained results, which further confirmed the deduction of fragmentation patterns and identified the compounds profile in Menthae Haplocalycis herba. Our research firstly systematically elucidated the non-volatile components of Menthae Haplocalycis herba, which laid the foundation for further pharmacological and metabolic studies. Meanwhile, our established method was useful and efficient to screen and identify targeted constituents from traditional Chinese medicine extracts.


Introduction
As there is growing interest in the use of traditional Chinese medicines (TCMs), systematic screening and identification of chemical components is essential for revealing the material basis of their therapeutic effects and ensuring their safety [1]. However, TCM extracts composed of multi-components are difficult to be comprehensively analyzed. Ultra-high performance liquid chromatography coupled with electrospray ionization tandem mass spectrometry (UHPLC-ESI-MS/MS) has been widely used as a powerful means for the analysis of multi-components in TCMs. Recently, B-ring from [M + H] + ion, which indicated they might be flavonones. The fragment ions at m/z 153, m/z 151 and m/z 137 were yielded by RDA cleavage, indicating there were two -OH on A-ring and one -OH and one -OCH3 on B-ring. Combining the literature data and Clog P values, F9 and F19 were identified as Homoeriodictyol and Hesperetin, respectively [14,15].
Compound F8 showed [M + H] + ion at m/z 611.19635 (C28H35O15, error < 5 ppm). The major fragment ions at m/z 465 and m/z 303 were yielded by successive neutral loss of Rha (146 Da) and Glc (162 Da) from [M + H] + ions. By comparing retention time, fragmentation behaviors with reference substance, it was unambiguously identified as Hesperidin [15,16].

Characterization of Phenolic Compounds
In the preliminary study, we found that UV absorption spectra of some peaks in the fingerprints of Menthae Haplocalycis herba were similar to salvianolic acid compounds. Meanwhile, by comparing with authentic standards and literature, PA5 and PA13 were identified as Rosmarinic acid and Salvianolic acid B, respectively [12,27,28].
Compounds PA6, PA12 and PA17 showed [M + H] + ions at m/z 539.11840 (C 27 H 23 O 12 , error < 5 ppm). After the CID cleavage, the further fragmentation of m/z 539 resulted in [M + H − C 9 H 10 O 5 ] + at m/z 341 and [M + H − C 9 H 10 O 5 − CO 2 ] + at m/z 297, involving the presence of tanshinol. By referring to literature data, the properties of Cis-salvianolic acid J, Lithospermic acid and trans-salvianolic acid J were in accordance with the description. Therefore, PA12 was unambiguously defined as Didymin by comparing with reference compound, while PA6 and PA17 were finally deduced as Cis-salvianolic acid J and trans-salvianolic acid J according to the literature data and Clog p values [12,26,27,29].

Characterization of Terpenoids
It is difficult to determine terpenoids by UHPLC-PDA analysis because of their weak UV absorption. UHPLC-MS/MS is a powerful technique to identify these kinds of constituents. In their ESI-MS/MS spectra, terpenoids usually lose a molecule of H 2 O or CH 3 because it normally contains hydroxy and methyl groups. For terpenoids glycosides, [M + H − 162] + was easily monitored as the characteristic ion by losing a dehydrated glucose. Moreover, fragment ions referred above often have high abundance. Based on these fragmentation pathways, four monoterpenoid aglycones, six monoterpenoids glycosides and two triterpenoids were finally identified.

Identification of Monoterpenoid Glycosides
With respect to compound T1, [M + Na] + adduct ion at m/z 369.15198 (C 16 H 26 O 8 Na, error < 5 ppm). After the CID cleavage, m/z 351, m/z 328, m/z 207 and m/z 185 were produced by successively neutral loss of one molecular of water and dehydrated glucose. According to the above analysis, T1 was plausibly described as petroside [31]. Likewise, by comparing with the authentic compound, T6 was unambiguously identified as (1R,2R,4S)-trans-1, Similarly, based on the above analysis, T3, T4, T7 and T8 were tentatively deduced as (2R,3R [33]. Our data demonstrated that some monoterpenoids are the glycosides of volatile constiuents in Menthae Haplocalycis herba. Taking T8 for example, it is the glucuronide of menthol, as we all know that menthol is the main effective component attributed to volatile constituents in Menthae Haplocalycis herba. Meanwhile, some monoterpenoid glycosides may be metabolized by intestinal flora after hydrolysis of aglycones (probably volatile constituents) and then absorbed into blood to display pharmacological effects. In this sense, it will be of great significance to carry out the study of non-volatile constituents in Menthae Haplocalycis herba.

Characterization of Phenylpropanoids
In the ESI-MS/MS spectra, phenylpropanoids always lose a molecule of H 2 O because they contain hydroxy groups. For phenylpropanoids glycosides, [M + H − 162] + was easily monitored as their characteristic fragment ion. According to the fragmentation pathways, two phenylpropanoids glycosides were tentatively identified. Compounds

Target Isolation and Verification
Acacetin, Rosmarinic acid and Clemastanin A were obtained from the effluent fraction of H 2 O-MeOH (50:50→75:25, v/v) by multiple isolation means. Their structures were verified combined with the 1 H-NMR, 13 C-NMR, which consistent with bibliographies [34][35][36][37]. The obtained MS data [35][36][37] of them were in accordance with the deduction of the MS/MS results, which further demonstrated the reliability of the deduced fragmentation patterns and identified the profile of non-volatile constituents in Menthae Haplocalycis herba.

Materials and Reagents
HPLC grade acetonitrile and formic acid were supplied by Fisher Scientific (Fisher, Fair Lawn, NJ, USA). Ultrapure water was purchased from Hangzhou Wahaha Group Co., Ltd. (Hangzhou, Zhejiang, China). All of the other reagents and chemicals were of analytical grade and commercially available.
Dried Herbal medicine samples of Menthae Haplocalycis herba were purchased from Anguo Linshi Medicinal Materials Co., Ltd. in Hebei, China and were authenticated as the aerial part of Menthae. haplocalyx Briq, which was harvested in Jiangsu at autumn by Professor Chunsheng Liu at the Beijing University of Chinese Medicine (BUCM, Beijing, China). All Menthae Haplocalycis herba samples were stored in Chinese medicine institutes of BUCM.

Standard Solutions
Stock solutions were prepared by dissolving appropriate amounts of 10 reference compounds in methanol. Proper amounts of each stock solution were then transferred to a 25 mL volumetric flask, and then methanol was added to make up the volume to obtain a final mixed reference solution. All the solutions were stored at 4 • C and brought to room temperature before use.

Sample Solutions
Sample (0.2 g) milled by 65 meshes beforehand were extracted with 10 mL methanol in an ultrasonic bath for 30 min. After being cooled to room temperature, it was weighed and adjusted to the original weight by adding methanol, and then filtered through a 0.22 µm nylon filter for analysis.

UHPLC-ESI-MS/MS System
UHPLC-ESI-MS/MS analysis was performed on a DIONEX Ultimate 3000 UHPLC system (Thermo Fisher Scientific, Waltham, MA, USA) with a binary pump and an autosampler. A series of preliminary experiments were performed to optimize mobile phase composition and elution conditions. Finally, analysis was carried out at 35 • C on an Acquity UPLC ® BEH C 18 column (2.1 mm × 100 mm, 1.7 µm, Waters Corporation, Milford, MA, USA). The mobile phase consisted of 0.2% formic acid aqueous solution (A) and acetonitrile (B). A gradient program was adopted as follows: 0-5 min, 5%-19.5% B; 5-8.

Peak Selections and Data Processing
Thermo Xcalibur 2.1 workstation (Thermo Fisher Scientific, San Jose, CA, USA) was used for data acquisition and processing. In order to obtain as many fragment ions of non-volatile compounds of Menthae Haplocalycis herba as possible, the peaks detected with intensity over 30,000 were selected for identification. The chemical formulas for all parent ions of the selected peaks were calculated from the accurate mass using a formula predictor by setting the parameters as follows: C (0-50), H (0-100), O (0-30), Cl (0-2), N (0-2), Na (0-2), K (0-1) and ring double bond (RDB) equivalent value (0-20). Other elements such as Br and P were not considered because they are rarely present in Menthae Haplocalycis herba.

Extraction and Isolation of Target Compounds
The air dried Menthae Haplocalycis herba samples (20.0 Kg) were extracted three times with tenfold excess of 70% EtOH under reflux for 1.5 h each at 80 • C. The combined extract was evaporated under reduced pressure to obtain a crude residue. This residue was further dispersed in H 2 O, and then successively passed through a Dianion HP (Mitsubishi Chemical Co., Kyoto, Japan) 2MGL macroporous resin column and then washed with extracted with H 2 O-MeOH (5:95→MeOH, v/v). The H 2 O-MeOH (50:50→75:25, v/v) extract was further purified by multiple isolation methods, such as silica gel column chromatography, C 18 antiphase silica gel column chromatography, Sephadex LH-20 gel chromatography, HPD-400 macroporous resin column, etc. Acacetin, Rosmarinic acid and Clemastanin A were obtained finally.

Conclusions
Our study took advantage of the UHPLC-LTQ-Orbitrap HRMS system and firstly reported the identification of 64 non-volatile compounds with various structure types, including 30 flavonoids, 20 phenolic acids, 12 terpenoids and two phenylpropanoids in Menthae Haplocalycis herba. Finally, target isolation of three compounds named Acacetin, Rosmarinic acid and Clemastanin A were performed based on the obtained results, which further confirmed the deduced fragmentation patterns and identified the profile of non-volatile constituents in Menthae Haplocalycis herba. The results also clearly elucidated that there may exist some inevitable relations between volatile and non-volatile constituents. Meanwhile, our developed method has been shown to be an excellent tool for the systematic characterization of non-volatile constituents in Menthae Haplocalycis herba, which also benefits its further pharmacological research and clinical applications. Moreover, this study sets a good example for the rapid identification of chemical constituents in TCMs.