The integrated analysis of host metabolome and intestinal microbiome is an opportunity to explore the complex therapeutic mechanisms of traditional Chinese medicines. Currently, researchers mainly employ various statistical correlation analytical methods to investigate metabolome-microbiome correlations. However, these conventional correlation techniques often focus on statistical correlations and their biological meanings are always ignored, especially the functional relevance between them. Here, we developed a novel enzyme-based functional correlation (EBFC) algorithm to further improve the interpretability and the identified scope of microbe-metabolite correlations based on the conventional Spearman’s analysis. The proposed EBFC algorithm is successfully utilized to reveal the therapeutic mechanisms of Jian-Pi-Yi-Shen (JPYS) formula on the treatment of adenine-induced chronic kidney disease (CKD) rats. JPYS, a TCM formula for treating CKD, has beneficial clinical effects. We tentatively revealed the potential mechanism of JPYS for treating CKD rats from the perspective of the serum metabolome, gut microbiome, and their interactions. Specifically, 11 metabolites and 19 bacterial genera in the CKD rats were significantly regulated to approaching normal status after JPYS treatment, suggesting that JPYS could ameliorate the pathological symptoms of CKD rats by reshaping the disturbed metabolome and gut microbiota. Further correlation analysis between the significantly perturbed metabolites, microbiota, and the related enzymes provided more strong evidence for the study of host metabolism-microbiota interactions and the therapeutic mechanism of JPYS on CKD rats. In conclusion, these findings will help us to deeply understand the pathogenesis of CKD and provide new insights into the therapeutic mechanism of JPYS.