Mass spectrometry with paper spray ionization in analysis of the Arrabidaea chica leaf tea

The species Arrabidaea chica (Humb. & Bonpl.) B. Verl. (Bignoniaceae) is a plant native to tropical forests, popularly known as pariri or crajiru, and it is widely used in folk medicine to treat inflammatory diseases, anemias, skin diseases and to assist in the treatment of leukemia. Because of the widespread use of the plant, concern with its quality and effectiveness is fundamental. This study demonstrates the application of mass spectrometry with ionization by paper spray (PS-MS) as a chemical characterization method to evaluate the presence of the active principles in teas prepared by infusion from commercial samples of the medicinal plant Arrabidaea chica by infusion. The analyses of the samples were performed in two periods: 24 hours and 15 days after the preparation of the teas. Compounds referring to the class of substances 3-deoxythocyanidins were detected without any pre-treatment of the sample or chromatographic separation to provide valuable information for the evaluation and quality control of this product. Significant changes in the m/z signals of greater intensity, referring to the class of 3-deoxythocyanidins, were observed in the mass spectra, and multivariate analysis corroborated the degradation of this class of substances with time, resulting in a probable loss of the quality and effectiveness of the tea.


Introduction
The species Arrabidaea chica (Humb. & Bonpl.) B. Verl. belongs to the family Bignoniaceae. It is distributed mainly in the tropical regions of South America and Africa. In Brazil, it does not have a unique habitat, being found in several regions, mainly in the Amazon (Medeiros et al., 2011). The plant is popularly known as crajiru (Amazonas), carajiru, pariri (Pará), and vine-cross, among other names (Ferreira et al., 2013).
According to folk medicine, the leaves of Arrabidaea chica have anti-inflammatory, healing, anti-anemic and astringent properties. They are used to fight hemorrhage, leukemia and skin diseases (Kalil Filho et al., 2000;Behrens et al., 2012). The widespread use of the plant in medicine and its pharmacological activities has aroused the interest of several researchers. Thus, the standardization of its extracts through studies of its chemical profile is essential (Behrens et al., 2012;Schiozer et al., 2012).
Ambient ionization is an area of mass spectrometry that has emerged quickly in several methods, all aiming at direct and rapid sampling of analytes in complex samples with the minimum of previous treatment in their ambient state . Recently, Wang and collaborators (2010) described a new technique of ambient ionization called mass spectrometry with paper spray ionization (PS-MS).
The PS-MS ionization method is based on the generation of analyte ions by applying a high voltage on a chromatographic paper triangle with a small volume (<10 μL) of the sample, which is performed at the entrance of a mass spectrometer. The samples are loaded onto the paper with the addition of an appropriate solvent, which was usually methanol.
The applied potential difference causes the formation of a fine spray of solvent droplets with the analyte on the tip of the paper because of the greater accumulation of charges .
In the literature, the method has been approached in different matrices such as in the analysis of drugs in biological fluids (blood and urine) , analysis of natural products such as plants and teas (Liu et al., 2011;Deng and Yang, 2013), forensic sciences (Ferreira et al., 2015), food (Soparawalla et al., 2011;Zhang et al., 2012;Garret et al., 2013) and discrimination of microorganisms (Hamid et al., 2014). This work sought to evaluate and compare the chemical fingerprint profile of teas obtained by infusing Arrabidaea chica leaves from different commercial samples and to verify the influence of time on its composition using mass spectrometry with PS-MS.

Methodology
The work corresponded to laboratory activities, with the extraction by infusion of pariri leaves and the identification of its constituents, after 24 hours and 15 days of extraction, as well as the study of grouping these extracts, considering the time and the bioactive compounds present. Furthermore, it was characterized as a descriptive research because, according to Oliveira (2011), it took into account the observation, registration and analysis of the objects of study and their relationship with other phenomena.

Plant material
A total of fourteen different samples from the Arrabidaea chica plant were purchased from different commercial establishments in Belo Horizonte (MG) and the metropolitan region in the period from February to March 2018.

Preparation of extracts
The extracts were prepared by infusing 1.0 g of the dried leaves of the samples of Arrabidaea chica in 30 mL of distilled water at 100 ºC for a period of one hour. The extracts were filtered and stored in amber bottles under refrigeration at 6 to 10 ºC

Mass Spectrometry with Paper Spray Ionization
For the paper spray ionization analysis, the chromatographic filter papers were cut into triangles (10 mm high and 10 mm base) and attached to a metal alligator clip at 5 mm from the mass spectrometer source. About 10 µL of the tea samples were transferred directly to the paper triangle and left to dry at room temperature. Then, 10 µL of methanol were added and voltage (3 kV) was applied to obtain the mass spectra. The analyzes were performed on teas prepared 24 h (A) and 15 days (B) after the extracts were prepared.

Instrumentation
All mass spectra were obtained using a Thermo Fisher LCQ FLEET mass spectrometer with a low-resolution ion trap mass analyzer, operating in the positive ion mode. The ionization source was built in the laboratory with a three-dimensional movement to adjust the paper with the mass spectrometer input. The spray voltage was adjusted to 3 kV, the capillary temperature was adjusted to 275 ºC and a 40 V potential, and the potential of the lenses was 120 V. For PS-MS, mass spectra were acquired in positive ion mode over m/z range 250-350.

Multivariate statistical analysis
The differentiation of the extracts was achieved through Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA). The data were processed in the Minitab 19.2 software. A data matrix with 28 lines (14 commercial samples in periods A and B) and 20 columns (the relative intensities values of the most abundant mass spectra signals) was used. The PCA was performed using a covariance matrix, and the scores of the main components (cumulative eigenvalue greater than 80%) were used for the construction of the dispersion graph to evaluate the possible groupings of the commercial samples. In turn, HCA was performed using the Euclidean distance measurement, non-standard data and the centroid linkage method. The respective dendrogram was evaluated for possible groupings of the commercial samples.

Results and Discussion
The choice for the positive ion mode in the analyses was based on the cationic characteristic of the 3deoxyanthocyanidins (1-4). The ions m/z 285, 299, 301 and 315 were identified in the sample of A. chica in previous analyses by ESI-(+)-MS (Schiozer et al. 2012, Mafiolletti et al., 2013, Taffarello et al., 2013. In addition to these four substances, Cabral (2010)    The use of multivariate analysis of the A. chica extracts was able to grouped the samples based on the differences in the spectra: 24 hours after (Samples A) and 15 days after (Samples B). The Principal Component Analysis with covariance matrix was able to explain 86.3% of the total data variability. The Figure 3 represents the first two components because the representation in three dimensions does not include new behaviors.  Cluster analysis ( Figure 5) was also performed, and the same grouping was observed for the centroid link method with Euclidean distance measurement, considering the original non-standard data. The four groups were formed with a similarity level of 72.74%. Source: Authors.
The same pattern of four groups could be seen in the results, considering the 25, 20 or 15 most abundant signals of the mass spectra, while even maintaining the greater homogeneity of samples A.

Conclusion
Through the analysis of the mass spectra of the leaf teas of the samples acquired in different commercial establishments in Belo Horizonte, MG, and metropolitan region, it was concluded that the samples were obtained from the species Arrabidaea chica. Being possible to identify, the chemical markers of this species related to the class of substances 3 deoxyanthocyanidins (m/z 285, 299, 301, and 315) were identified.
Significant changes in the m/z signals of greater intensity were observed for the different sample reading conditions (24 hours and 15 days after the preparation of the leaf teas). The instability of these classes of substances with time and, consequently, the loss of these active ingredients, which result in changes in the quality and effectiveness of the tea, could be seen. This fact shows that the tea should be used immediately after its preparation.
Ultimately, the experimental data demonstrated that the chemical characterization by PS-MS is a simple, fast, and robust method for the aqueous plant extracts evaluation, having potential for future work on quality analysis and standardization of aqueous extracts of importance to the food, beverage, and agricultural industries, and in the health area.