Food preferences are shaped by various factors, including cultural background, which significantly influences an individual's exposure to and acceptance of particular foods. Wine, as a complex beverage with diverse aroma compounds, provides a suitable model for studying the impact of culture and individual physiology on food perception. Previous studies have highlighted the role of saliva in modifying wine aroma through interactions with volatile compounds, suggesting a potential link between salivary composition and wine preference. Saliva contains various enzymes (e.g., esterases, α-amylase) and proteins that can influence aroma release through hydrolysis, binding, and other interactions. Furthermore, cultural differences in diet and lifestyle might lead to variations in salivary parameters, including protein composition and microbiota. This study aimed to investigate the hypothesis that differences in salivary protein composition between experienced Western and Chinese wine tasters explain variations in their wine perception and preference. The study's significance lies in potentially revealing a biochemical basis for understanding food preferences and tailoring food products to specific cultural groups.

Existing literature indicates a strong influence of cultural background on food preferences, with geographic and climatic conditions influencing food availability and shaping culinary traditions. The perception of wine aromas, both orthonasal and retronasal, is crucial for wine quality rating, and saliva plays a significant role in shaping the retronasal aroma profile. Salivary enzymes and proteins interact with wine volatiles, affecting their release and perception. Studies have shown that factors like psychological stress, smoking, caffeine intake, exercise, and diet alter salivary parameters. Ethnic differences in salivary composition have been noted, with variations in enzyme activities linked to distinct dietary habits. Previous research has indicated that differences in salivary microbiome composition might contribute to differential perception of wine attributes like bitterness and astringency. While some studies have reported differences in salivary protein concentration between ethnic groups, a comprehensive proteomic analysis was lacking. The impact of saliva on aroma release has been demonstrated *ex vivo*, showing differences in aroma retention related to saliva protein content and antioxidant capacity. Previous sensory studies revealed cultural differences in wine preferences, but a direct link to salivary composition remained unexplored.

The study recruited 13 experienced Western and 13 experienced Chinese wine tasters, matched for gender as much as possible, to minimize the impact of wine education and cultural biases on sensory evaluation. Eight Shiraz wines from various regions in Victoria, Australia, were selected, varying in chemical parameters (pH, alcohol, residual sugar, total phenolic and tannin content). The sensory evaluation involved two sessions: (1) Pivot Profile – a free-description-based method capturing qualitative descriptions of wine attributes and (2) Continuous Scale Assessment – quantifying perceived intensities of various sensory attributes and overall liking. Before the sensory evaluation, saliva samples were collected using a standardized protocol. Chemical analysis of the wines involved determining total residual sugar, alcohol content, total titratable acidity, pH, volatile acid content, malic acid content, total phenolic content (TPC), and total condensed tannin content (CTC) and tannins using bovine serum albumin (BSA) precipitation. Saliva samples were analyzed for total protein concentration, α-amylase activity, and esterase activity. Comprehensive two-dimensional gas chromatography-mass spectrometry (GC × GC-MS) analyzed the headspace volatiles of wines spiked with pooled saliva samples from both groups to assess the impact of saliva on volatile release. Tandem Mass Tag (TMT) quantitative proteomics analyzed the salivary protein composition. Statistical analyses included principal component analysis (PCA) and correspondence analysis (CA) for sensory data, t-tests for comparing protein concentration and enzyme activities between groups, Wilcoxon Signed Rank test for comparing perceived intensities, and Pearson correlation analysis to investigate relationships between salivary protein concentrations and perceived intensities of wine attributes. A supplementary analysis assessed the impact of β-lactoglobulin (BLG), a lipocalin protein, on wine volatile release using a similar method as the pooled saliva spiking.

Sensory evaluation revealed significant differences in wine flavor perception between Western and Chinese tasters. The Chinese group perceived stronger fruity, floral, and sweet notes in several wines. PCA and CA of sensory data showed distinct patterns in attribute perception and overall liking between the groups, with the Chinese group showing a preference for wines with earthy and umami characteristics and higher alcohol, residual sugar, and condensed tannin content, while Western tasters preferred wines with floral notes. GC × GC-MS analysis revealed significant differences in wine volatile release after spiking with pooled saliva. The Chinese pooled saliva resulted in higher concentrations of most volatiles (esters, alcohols, aldehydes), while the Western pooled saliva increased only a few specific compounds. While there were no significant differences in total salivary protein concentration, α-amylase, and esterase activity between the two groups, significant variations were found in specific proteins. Proteomic analysis identified 35 proteins significantly more abundant in the Chinese group and 3 proteins significantly more abundant in the Western group. Among these, lipocalin-1 (LCN-1) and proline-rich proteins (PRPs) showed correlations with perceived wine attributes. PRPs showed negative correlations with perceived floral and fruity notes. LCN-1 showed both positive and negative correlations with fruity and floral notes depending on the wine sample. A supplementary experiment using BLG did not show a significant impact on aroma compounds, suggesting LCN-1's role might be indirect or require further investigation.

The findings support the hypothesis that variations in salivary protein composition contribute to differences in wine perception and preference between cultural groups. Differences in volatile release, as detected by GC × GC-MS, align with the sensory perception results. The variations in PRPs and LCN-1 concentrations likely mediate the observed differences in aroma perception. The negative correlation between PRPs and fruity/floral notes is likely due to PRPs’ ability to bind tannins, forming complexes that sequester volatile aroma compounds. The role of LCN-1 remains unclear, with findings suggesting a possible indirect mechanism or a limited capacity to directly influence the volatile profile due to its low concentration in saliva. The study provides evidence of a biochemical basis for understanding food preferences and opens avenues for targeted food product development.

This study demonstrates that salivary protein composition, particularly the abundance of PRPs and LCN-1, influences wine perception and preference across cultures. These findings offer valuable insights into the biological underpinnings of food preferences and highlight the potential for utilizing salivary proteomics to understand and tailor food products to specific consumer groups. Future research could focus on elucidating the precise mechanism by which LCN-1 affects aroma perception, investigating the influence of individual dietary factors on salivary protein composition and exploring the broader implications of salivary proteomics in food science and personalized nutrition.

The study used a relatively small sample size (n=26), which might limit the generalizability of the findings. The study focused on experienced wine tasters, potentially excluding the preferences of less experienced consumers. The study was limited to Shiraz wines from a specific region; different wine types and regions might exhibit different patterns. The exact mechanism by which LCN-1 influences aroma perception remains to be fully elucidated. Further research using larger and more diverse populations is needed to validate the findings and explore the full spectrum of factors contributing to variations in food preferences.