An origin identification model for labeling of shiitake (*Lentinula edodes*)

The globalization of food markets necessitates reliable geographical identification to combat food mislabeling. While origin labeling is established for products like cheese and meat in the EU and Japan, a standardized system for non-soil-based agricultural products like mushrooms is lacking. Shiitake mushrooms (*Lentinula edodes*), the second most popular mushroom globally, are increasingly produced using a sawdust block method. In Korea, shiitake mushrooms originate from domestic sources, imported Chinese inoculated sawdust blocks, and China itself. This leads to origin labeling challenges, particularly with mushrooms cultivated in Korea using imported Chinese inoculated sawdust blocks. Current Korean labeling rules (Origin Labeling Act, implemented December 2020) depend on the duration of sawdust block preparation, inoculation, and cultivation. The World Trade Organization guidelines also play a role, specifying labeling based on the country of final substantive transformation. This creates ambiguities and controversies among producers. While various analytical techniques exist for food origin identification, few are dedicated to mushroom species. Current identification methods in Korea mainly rely on visual inspection of fruiting body morphology, lacking analytical rigor. Stable isotope ratio analysis (SIRA), considering factors like δ¹³C, δ¹⁵N, δ¹⁸O, and δ³⁴S, offers a promising approach due to the unique isotopic signatures of organisms based on environmental and geological factors. However, its application to non-photosynthetic, cultivated mushrooms grown on substrates requires investigation, especially considering how the cultivation substrate and environment influence isotopic fractionation. This study aims to evaluate SIRA's feasibility in determining the geographical origin of shiitake mushrooms consumed in Korea, specifically addressing the complexities arising from the use of imported sawdust blocks.

Existing literature highlights the importance of geographical origin labeling in food trade and consumer protection. Several studies have used SIRA combined with chemometrics to identify the origin of various food products, including cereals, beverages, vegetables, meats, and wines. However, limited research exists on the origin identification of mushroom species. Previous studies have attempted to differentiate shiitake mushrooms based on stable isotopes (δ¹³C and δ¹⁵N), but these efforts have not fully addressed the complexities of sawdust block cultivation and the origin of the cultivation substrate. This study aimed to bridge this knowledge gap by developing a robust model for shiitake origin identification, considering the multiple sources of mushrooms available in the Korean market.

Fresh shiitake mushrooms (1 kg) were collected from Korean farms and markets between 2017 and 2019. Samples were categorized into three groups: Korean origin (domestic sawdust blocks), Chinese inoculated sawdust blocks (Chinese blocks used in Korean cultivation), and Chinese origin (fully produced in China). Shiitake grown via traditional log cultivation were also included for comparative analysis. Samples were lyophilized, pulverized, and prepared for stable isotope ratio analysis (SIRA). δ¹³C, δ¹⁵N, δ¹⁸O, and δ³⁴S were measured using various elemental analyzers and isotope ratio mass spectrometers. Data analysis involved least significant difference tests (general linear model), discriminant analysis (DA) with Wilks's Lambda method, and receiver operating characteristic (ROC) curves. The DA models were evaluated for classification accuracy using the leave-one-out cross-validation method. Different models were created, treating Chinese inoculated sawdust blocks as either Korean or Chinese origin to assess the impact of different origin assignment scenarios on model accuracy.

Significant differences in δ¹³C, δ¹⁵N, δ¹⁸O, and δ³⁴S were observed among shiitake mushrooms based on cultivation method (log vs. sawdust block) and geographical origin of sawdust blocks. Sawdust block cultivation showed higher δ¹³C, δ¹⁵N, and δ³⁴S values, but lower δ¹⁸O compared to log cultivation. Two-dimensional plots revealed that δ¹⁵N provided the clearest separation between cultivation methods. A discriminant function using all four isotopes (δ¹³C, δ¹⁵N, δ¹⁸O, δ³⁴S) achieved high classification accuracy (97.7%) for distinguishing between log and sawdust block cultivation methods. However, when attempting to discriminate the geographical origin of sawdust blocks (Korean, Chinese inoculated, Chinese), the model achieved lower accuracy (78.4% original, 77.7% cross-validated). The model frequently misclassified mushrooms of Chinese origin as originating from Chinese inoculated blocks. A revised model, treating Chinese inoculated sawdust blocks as Chinese origin, significantly improved accuracy (93.5% for both original and cross-validated datasets). In this improved model, δ¹⁵N was the most crucial predictor variable for determining the origin of the mushrooms. Conversely, when the Chinese inoculated blocks were treated as Korean, the accuracy dropped significantly (82.0% original set, 80.9% cross-validated).

This study demonstrates the feasibility of using SIRA for geographical origin identification of shiitake mushrooms, particularly considering the complexities of sawdust block cultivation. The high accuracy achieved in distinguishing between cultivation methods highlights the isotopic differences resulting from different nutrient availability and environmental conditions. The lower accuracy in discriminating the geographic origins of sawdust blocks emphasizes the need to consider the source of the substrate material when determining the origin of the final product. The improved accuracy obtained by treating Chinese inoculated blocks as Chinese origin aligns with the principle of considering the origin of the substrate material. The findings challenge the current Korean origin labeling system, which allows labeling as 'Korean origin' even when using imported Chinese inoculated sawdust blocks. The study suggests the need for more transparent labeling practices that reflect the origin of both the substrate and the cultivation location. The approach used could be extended to other mushroom species and countries with similar import practices.

The study highlights the limitations of current origin labeling systems for cultivated mushrooms. The developed SIRA-based model provides a reliable method for identifying the geographic origin of shiitake mushrooms, especially when considering the source of the sawdust blocks. The results underscore the need for revised labeling guidelines that accurately reflect the origin of the substrate and cultivation location to ensure fair trade practices and avoid mislabeling. Future research could investigate other isotopic markers or combine SIRA with other analytical techniques for enhanced accuracy and to expand this method to other mushroom varieties and geographic regions.

The study's findings are based on samples collected from a specific region (Korea) and time period. The generalizability to other geographic locations and time periods might require further validation. The model's accuracy depends on the assumption of the origin of the Chinese inoculated sawdust blocks, requiring standardized procedures for assigning origin in such cases. Finally, the study did not consider potential variations in isotopic signatures introduced during transportation and processing of the mushrooms.