Differential colorimetric nanobiosensor array as bioelectronic tongue for discrimination and quantitation of multiple foodborne carcinogens

• Published in: Food chemistry Vol. 357; p. 129801
• Main Authors: Fang Wong, Siew, Mei Khor, Sook
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 30.09.2021
• Subjects: Acrylamide; Differential sensing; Hierarchical cluster analysis; Nanoparticle−protein interactions; Principal component analysis; Nanoparticle−protein interactions; Acrylamide; Differential sensing; Hierarchical cluster analysis; Principal component analysis
• ISSN: 0308-8146; 1873-7072
• DOI: 10.1016/j.foodchem.2021.129801

[Display omitted] •Differential colorimetric nanobiosensor array was developed for detection and discrimination of acrylamide and six analogues.•Albumins were introduced as cross-reactive receptors for differential sensing.•Target analytes were distinguished based on their amine subgroups, IARC classifications, and food additive types.•Non-targeted analytes were identified and differentiated by sweetener and food ingredient types.•PLSR model was applied to predictively quantify the analyte concentration in unknown coffee samples. Foodborne amides, specifically acrylamide, are vitally important for food safety and security, as they are the most common food toxicants and suspected human carcinogens. A facile and novel differential-based colorimetric nanobiosensor array composed of three surface-bioengineered gold nanoparticles (AuNPs) was developed for the rapid detection, differentiation, and quantification of acrylamide and six analogues. Diverse cross-reactive receptors demonstrated differential binding affinities toward target analytes, resulting in distinctive AuNP aggregation behaviors and distinguishable response patterns. The sensor array, integrated with principal component analysis and hierarchical cluster analysis, accurately discriminated foodborne amides based on their amine subgroups, International Agency for Research on Cancer (IARC) carcinogen classifications, and food additive types, even at ultra-low concentrations (500 pM). Additionally, the sensor array successfully differentiated non-targeted analytes by sweetener and food ingredients types with 100% correct classification. Partial least squares regression outcomes exhibited high correlation coefficients (R2 > 0.95). Thus, the sensor array has practical potential for food safety monitoring in the food and beverage industries.