Application of HPLC Fingerprint Coupled with Chemical Pattern Recognition and Multicomponent Quantification for...

Large amounts of banana leaves are discarded as agricultural waste during banana harvesting; however, they are rich in bioactive compounds with promising exploitation potential. The ethyl acetate fraction contains the highest levels of total phenolics and total flavonoids together with strong antioxidant activity, which renders it a candidate source of natural antioxidants. In our current study, HPLC fingerprints were constructed for ten batches of ethyl acetate extracts from banana leaves (BLEA), and fingerprint similarity evaluation coupled with statistical analysis of common peak areas was carried out. Meanwhile, an HPLC quantitative method was developed to simultaneously determine eight characteristic components. The similarity values between the fingerprints of ten batches and the reference fingerprint 'R' ranged from 0.791 to 0.988. Cluster analysis (CA) and orthogonal partial least squares discriminant analysis (OPLS-DA) yielded consistent results, revealing clear separation trends among samples from different batches. In this quantitative method, eight characteristic analytes showed favorable linearity within their respective concentration ranges (r ≥ 0.9997). The average recoveries ranged from 101.6% to 104.3%, with RSD of 0.32% to 0.84%, demonstrating that the quantitative method was accurate and reliable. HPLC fingerprint combined with multi-component content determination offers a scientific basis for the quality evaluation and biomass valorization of banana leaves.