Novelty application of multi-omics correlation in the discrimination of sulfur-fumigation and non-sulfur-fumigation Ophiopogonis Radix

Abstract

A rapid and sensitive approach to differentiate sulfur-fumigated (SF) Ophiopogonis Radix based on Multi-Omics Correlation Analysis (MOCA) strategy was first established. It was characterized by multiple data-acquisition methods (NIR, HPLC, and UHPLC-HRMS) based metabonomics and multivariate statistical analysis methods. As a result, SF and non-sulfur fumigated (NSF) Ophiopogonis Radix samples were efficaciously discriminated. Moreover, based on the acquired HRMS data, 38 sulfur-containing discriminatory markers were eventually characterized, whose NIR absorption could be in close correlation with the discriminatory NIR wavebands (5000-5200 cm^-1) screened by NIR metabonomics coupled with SiPLS and 2D-COS methods. This results were also validated from multiple perspectives, including metabonomics analysis based on the discriminatory markers and the simulation of SF ophiopogonin D and Ophiopogonis Radix sample. In conclusion, our results first revealed the intrinsic mechanism of discriminatory NIR wavebands by means of UHPLC-HRMS analysis. Meanwhile, the established MOCA strategy also provided a promising NIR based differential method for SF Ophiopogonis Radix, which could be exemplary for future researches on rapid discrimination of other SF Chinese herbal medicines.

Figure 1

Plant characteristics of Ophiopogonis Radix.

Figure 2

Summary diagram of the developed strategy and methodology.

Figure 3

The discriminatory information in the raw spectra (Fig. 3A) and preprocess method of SG9+1st (Fig. 3B) for PLS-DA model.

Figure 4

(A) The PLS-DA model of Ophiopogonis Radix using the UHPLC-HRMS data; (B) The PLS-DA model was considered as valid significantly since the corresponding Q2-intercept value is negative; (C) S-plot based on the UHPLC-HRMS data.

Figure 5

The mass fragmentation behaviors of identified markers (A) HRMS¹ spectrum of No. 11; (B) ESI-MS² spectrum of No. 11; (C) HRMS¹ spectrum of No. 20; (B) ESI-MS² spectrum of No. 20).

Figure 6

(A) The discrimination information of PLS-DA model combined by the preprocessing method of wavelet denosing of spectra; (B) The specific wavebands screened by the SiPLS; (C) The synchronous spectra of NSF and SF Ophiopogonis Radix. (D) PLS-DA model of Ophiopogonis Radix using the identified biomarkers data; (E) The PLS-DA model was considered as valid significantly since the corresponding Q2-intercept value is negative; (F) The synchronous spectra of NSF and SF of Ophiopogonin D; (G) Log (1/R)-Time curve of NIR analysis; (H) Total peak areas of these elucidated 38 compounds - Time curve.