Objective Uric acid (UA) and high-density lipoprotein (HDL) subclasses are associated with inflammation, coronary heart disease, and metabolic syndrome (MetS). However, the relation between UA and HDL subclasses is not well understood. Methods Subjects included 848 Japanese men not taking medication for hyperuricemia, hypertension, diabetes mellitus, dyslipidemia, or chronic renal disease; they underwent an annual health examination that included HDL subclass analyses. Results When subjects were stratified by HDL2-C or HDL3-C levels, UA level decreased as HDL2-C level increased, while UA levels increased as HDL3-C levels increased. In a multiple linear regression analysis, age, waist circumference (WC), diastolic blood pressure (BP), logarithmic transformed triglyceride ln(TG) and HDL3 cholesterol (HDL3-C), were associated with UA level. In a multiple logistic regression analysis for upper tertile of UA (≥ 6.8 mg/dL), WC, diastolic BP, ln(TG), HDL2-C and HDL3-C were associated. Since this analysis indicated that MetS components were determinants of UA level along with HDL-C subclass, possible synergistic effects of HDL-C subclass and MetS components to determine UA level were assessed. A combination of the number of MetS components and stratification of HDL3-C affected UA levels; the mean UA level increased in subjects with increased MetS components and HDL3-C level. Conclusion HDL-C subclasses were associated with UA level; particularly, a high HDL3-C level was associated with high UA level related to MetS in Japanese men.