Detection and Quantitation of Frauds in the Authentication of Cranberry-Based Extracts by UHPLC-HRMS (Orbitrap) Polyphenolic Profiling and Multivariate Calibration Methods.

UHPLC-HRMS (Orbitrap) polyphenolic profiling was applied to the characterization, classification, and authentication of cranberry-based natural and pharmaceutical products. Fifty three polyphenolic standards were characterized to build a user-accurate mass database which was then proposed to obtain UHPLC-HRMS polyphenolic profiles by means of ExactFinder software. Principal component analysis results showed a good sample discrimination according to the fruit employed. Regarding cranberry-based pharmaceuticals, discrimination according to the presentation format (syrup, sachets, capsules, etc.) was also observed due to the enhancement of some polyphenols by purification and preconcentration procedures. Procyanidin A2 and homogentisic, sinapic, veratric, cryptochlorogenic, and caffeic acids showed to be important polyphenols to achieve cranberry-based products discrimination against the other studied fruits. Partial least-squares regression allowed the determination of adulterant percentages in cranberry-fruit samples. Very satisfactory results with adulteration quantification errors lower than 6.0% were obtained even at low adulteration levels.


INTRODUCTION
Food manufacturers and society are concerned about food product quality. 26 Foodstuffs are complex products including, mainly, naturally occurring substances, but 27 other compounds such as those migrating from packaging materials or those coming 28 from technological and agrochemical processes can also be present. Typically, 29 organoleptic and socioeconomic factors influence foodstuff consumer preferences. 30 However, nowadays the presence of bioactive substances with healthy effects is gaining 31 interest in the society. Unfortunately, fraudulent practices derived from food product 32 adulterations by substitution, for instance, of the most valued components for others of 33 lower commercial value and lower health beneficial properties are being employed to 34 reduce food production costs. 1 For example, the addition of a co-fruit (a more economic 35 and accessible fruit) to the final fruit-based processed foodstuffs such as juices is among 36 the most common fraudulent practices that can be found in the fruit industry. 2 Fruit-37 based pharmaceutical preparations are also susceptible of fraudulent practices.  conditions at 10%B to re-equilibrate the column. The mobile phase flow-rate was 300 156 µL/min, and the injection volume employed (in full loop mode) was 10 µL.

157
The UHPLC system was coupled to a Q-Exactive Orbitrap HRMS system  Cranberry-based pharmaceutical syrups, fruits and raisins were freeze-dried to obtain 197 completely lyophilized products (Telstar LyoQuest lyophilizer, Terrasa, Spain) 198 following the method described by Pardo-Mates et al. 3 Briefly, a 24 h gradient 199 temperature ramp from -80 ºC to room temperature, followed by 6.5 h at 40 ºC, was 200 employed for lyophilization.

201
Sample treatment was carried out following a previously described method with some dependences of the studied polyphenols with the fruit products.

236
The percentage of fruit-extract adulterants (grape, blueberry or raspberry 237 extracts) in the cranberry-based extracts was quantified by PLS. Samples available were 238 distributed among training (calibration) and test (validation and prediction) sets (Table   239 1S in the supporting information). For both training and test steps, X-data matrices  Figure 1a). As can be seen in Table 2 example the MS/HRMS spectrum of rutin in Figure 1b).

271
Because of the great variety of chemical structures among the studied 272 polyphenols (see Table 1), MS/HRMS spectra were obtained by a data dependent 273 acquisition mode based on product ion scan applying for the fragmentation stepped interesting behavior was observed with the analyzed cranberry pharmaceutical samples.

335
Those manufactured as sachets and syrups were grouped together with cranberry-fruit 336 samples, while raw cranberry pharmaceutical extracts and capsules were completely 337 discriminated and perfectly separated.

338
To better study this behavior and taking into consideration the raspberry, 339 blueberry and grape extracts are expected to be used as adulterants of cranberry extracts, 340 as previously commented in the introduction section, independent PCA models between 341 cranberry-based samples and the other three fruit families studied were evaluated.
342 Figure 4 shows the score and loading plots of (a) PC1 vs PC2 for cranberry-and  The applicability of UHPLC-HRMS polyphenolic profiles for the authentication 384 and quantitation of fraud levels of adulterant fruit extracts by PLS was also evaluated.

385
For that purpose, cranberry-fruit extracts were adulterated with extracts of the other 386 three fruits (blueberry, raspberry and grapes) at different concentration levels (2, 2.5, 5, 387 6, 7, 12, 20 and 50%). Triplicates of all the adulterations as well as of 100% pure fruit 388 extracts were prepared. 50% adulteration was prepared in quintuplicate to evaluate data prediction errors obtained in all the adulteration cases studied are given in Table 3. As