Authenticating Botanicals and Herbal Supplements Using UPLC-QTof-MS and a Metabolomics Approach

Library Number:
Dr. Giorgis Isaac, Principal Scientist Scientific Operations, Waters Corporation
Content Type:
On Demand Webinar
Related Products:

This on-demand webinar presents various botanical studies (Chamomile and Hoodia) using UPLC-QTof MS metabolomics for the authentication of botanicals and their herbal supplements sold in the market. The metabolite profiles of the botanicals (Chamomile and Hoodia) and their commercial products were investigated using new informatics software to determine the pattern and identification of different metabolites. With this software, differential analysis of results across several botanical species can quickly be performed, thereby facilitating identification and quantitation of potential metabolite markers. Finally, significantly changing metabolite markers that differentiate between various botanical species were identified that can be used as a target marker for botanical authentication.

Key Learning Objectives

  • Develop multivariate approaches to authenticate commercial herbal supplements in the market.
  • Learn how to implement UPLC-QToF MS for the authentication of botanicals and their herbal supplements sold in the market.
  • Identify potential marker compounds that differentiate between botanicals and commercial herbal supplements using non-targeted analysis.

Watch the on-demand event here:


A large number of questions were asked by participants following the webinar, but many went unanswered due to time constraints. Please find those questions and answers below.

Q. What way of metabolomics may be more appropriate for herbal particular compounds, let's say anthocyanins and their post-metabolites in a target organisms (tissues), targeted or untargeted?

A. If you have known target compound classes in a particular tissue, then targeted metabolomics is suggested.

Q. What if compound form only sodium adducts, and if there is no other adducts formed?

A. The feature will be reported as m/z ion.

Q. PCA showed tea looked different from both Germen and Roman M. Could this be partially due to processing of tea?

A. Most of the tea samples are close to German Chamomile. Yes, for some of the tea samples the differences may be due to processing methods used in preparation of the tea samples.

Q. For an extracted compound, during database search, do we have to check Sodium and Proton adducts OR just Proton adducts, since the extracted compounds are predefined with Neutral Mass, RT, and m/z?

A. Depends on how the sample was prepared. If you choose only to check for proton adducts then any Na adducted forms of the compound will not be deconvolved with the protonated form thereby leading to an under estimation of the amount of the compound present in the sample. But if you know you are expecting to form sodium adducts in your sample, I would put them in as well for higher chance of identification.

Q. Regarding the MS preprocessing specially the alignment step of the retention time, as I understood, you are doing peak picking by contrasting each group against other and then proceed with the data process.

A. Following alignment as part of the peak picking process an additional aggregate run is generated from all of the aligned runs and a single peak picking is performed on this aggregate run. The pattern of picked peaks is passed to all the aligned runs ensuring that the same peak picking is performed on all the runs in the experiment.

Q. How long do you have to work with chromatographyc or the purity of chromatographic separation not so important in metabolomic analysis?

A. Chromatographic separation of compounds is important because some compounds may have similar masses and if not separated, we may miss them. The time depends on the number of compounds to be separated and the complexity of the sample.

Q. Is Progenesis QI Software compatible with other HRMS data like Orbitrap data?

A. Yes, Progenesis® QI software is a multi vendor mass spectrometry data processing software and compatible with the major mass spectrometry manufacturers.

Q. Would you please comment on the range of the polarity of compounds analysed?

A. The polarity of compounds analysed ranges from polar to non-polar. For example in Terminalia samples analysis, some compounds are highly polar like tannins (gallic acid, elagic acid etc) and some are non-polar like triterpenoids, based on these compounds polarity the separation becomes challenging.

Q. How do the database use retention time data as it was mentioned in the presentation two times.

A. The RT of a compound can be saved in the additional properties file in Progenesis QI or as a field in the database. When a search is performed where it is appropriate to include the RT as a property of the compound (i.e. the Chromatography is the same for the samples as it was when the database of compound Rts were generated) then the RT error can be used as a search criteria and will form part of the Identity Score.

Q. Why do you think that HRMS metabolomic fingerprint is more reliable for adulteration detection however the error of PLS regression is higher than classical methods?

A. HRMS metabolomic fingerprint is a powerful approach to use the full chemical profile to identify the samples compared to just using targeted compounds. We did not use PLS regression in this study but instead PCA to understand the identity of the commercial sample to the authenticated extracts. PLS regression is used more for quantifying the amount of a compound such as an adulterant in a sample but requires calibration samples first to build the model.

Q. Do you recommend adding internal standard or reference is sufficient?

A. In our case we performed untargeted metabolomics and used pooled quality control (QC) reference sample. Untargeted metabolomics is discovery driven/exploratory and there is no prior knowledge of the sample to be analyzed. QC reference is sufficient for untargeted metabolomics. But can include internal standard if you so wish and compare the abundances externally to the software. It is important to make sure the internal standard doesn't interfere with the sample or correlate with any particular marker compounds in the sample.

Q. Do you extract with only methanol?

A. Yes, in our case we used 100% methanol for LLE extraction. But 70/30 (methonal/water) is also very common extraction solvent for herbal supplement.

Q. What are QC pooled samples and what does it contain?

A. The quality control (QC) pooled samples were prepared by mixing 5 µL of each sample separately. Injecting QC sample between analytical runs ensures continuous quality assurance and provides confidence in your data. The QC samples can also be used for alignment and normalization of analytical experiments. Typically a QC sample can be a pooled sample or a standard test mix.

Q. In Progenesis IQ are we able to draw Venn diagrams between groups to get common compounds in comparison groups?

A. No, Venn Diagram option is not available in Progenesis QI. But Progenesis QI provides the facility to ‘Tag’ subgroups of compounds from the full compound list for further consideration or analysis. For example, you can create a tag to show all the compounds that are common in different groups. Q Do you have a database for packaging migrations? A. No, we do not have packaging migration database but you could use MetaScope to generate your own database based on relevant standards.

Q. Is it necessary to add internal standard in the samples?

A. No, it is not necessary to add internal standard in the sample but can include one if you so wish and compare the abundances externally to the software.

Q. Is it possible to miss some vital constituents (even markers) by methanol extraction process? If so, how to overcome this problem?

A. Most of the time we do not miss any marker compounds with methanol as an extraction solvent. Initially we do check the pure compounds solubility with different solvents (for wide range of compounds) and would provide an idea about the compounds solubility. Based on this the extraction solvent is chosen. In the beginning, the plant materials were also screened with different extraction solvents and check for the recovery of compounds of interest in case of targeted analysis.