Comparison of extracellular vesicles carrying bacterial DNA in urine and serum from a Korean population

Ga-Eun Yie¹Nam-Eun Kim²Sang-Cheol Park³Kangjin Kim⁴Sangah Shin⁵Seung Ku Lee⁶Chol Shin⁶Kwang Jun Lee^7*Sungho Won^8*

• ¹Interdisciplinary Program of Bioinformatics, Seoul National University, Seoul, Republic of Korea
• ²Institute of Health and Environment, Seoul National University, Seoul, Republic of Korea
• ³Institute for Biomaterials, Korea University, Seoul, Republic of Korea
• ⁴Department of Applied Statistics, Gachon University, Seongnam, Republic of Korea
• ⁵Department of Food and Nutrition, Faculty of Food Science and Technology, College of Biotechnology and Natural Resources, Chung-Ang University, Anseong, Gyeonggi, Republic of Korea
• ⁶Institute of Human Genomic Study, College of Medicine, Korea University, Seoul, Republic of Korea
• ⁷Division of Zoonotic and Vector Borne Diseases Research, Center for Infectious Disease Research, National Institute of Health, Korea Disease Control and Prevention Agency, Cheongju, Republic of Korea
• ⁸Seoul National University, Seoul, Republic of Korea

Bacteria-derived extracellular vesicles (BEVs) are emerging as key biomarkers of hostmicrobiota interactions. However, little is known about how BEV profiles differ across different biofluids or how these differences relate to clinical phenotypes. We aimed to examine the BEV distribution and site-specific and shared associations with host phenotypes, and evaluated the clinical relevance of microbial distance between sampling sites. In this study, we profiled BEVs using 16S rRNA sequencing of urine and serum samples from middle-aged and older Koreans (n = 2,827). We compared the alpha and beta diversities between the two biofluids, and assessed their relative abundances and associations with host anthropometric measurements, blood tests, and dietary nutrient intake. We also calculated the distances between urine and serum BEV compositions and assessed their clinical and metabolic implications. The results revealed that the urine BEVs exhibited higher alpha diversity than the serum BEVs, as well as stronger associations with dietary nutrient intake, particularly sugar, and with anthropometric measures such as waist circumference. The correlations between the urine and serum BEV compositions were generally low, emphasizing their distinct microbial profiles. Notably, individuals with shorter urine-serum BEV composition distances had higher waist-to-hip and sugar-to-fat ratios. This study provides a comprehensive comparison of urinary and serum BEVs, revealing the differences in microbial composition and associations with host phenotypes. In particular, urine BEVs showed stronger associations with dietary and metabolic factors, underscoring their potential utility as non-invasive biomarkers for metabolic health.