**Background:** Chronic inflammation is a key driver of colon carcinogenesis, and the gut microbiome plays a critical role in inflammation-related colorectal cancer. Perilla frutescens (PF) leaves have demonstrated antioxidant, anti-inflammatory, and anti-cancer activities, but the bioactivity of PF seed residue—a waste product from seed oil production—remains largely unexplored. This study aimed to evaluate the chemopreventive potential of PF seed residue crude ethanol extract (PCE) on the promotion stage of colon carcinogenesis in a rat model, focusing on inflammatory processes and gut microbiota modulation.
**Methods:** Dried PF seed residue was extracted with 70% ethanol to obtain PCE powder. Total phenolic and flavonoid contents were measured colorimetrically, and key phenolic compounds (rosmarinic acid, luteolin, apigenin) were quantified by HPLC. For the in vivo ACF progression study, 30 male Wistar rats were randomly divided into five groups (n=6 each). Groups 2–4 received subcutaneous injections of 40 mg/kg dimethylhydrazine (DMH) at weeks 1 and 2, followed by 1% dextran sulfate sodium (DSS) in drinking water for one week (week 3) to induce colonic inflammation and promote ACF formation. From week 5 to week 15, rats were orally gavaged daily with 10% DMSO (group 2, positive control), 0.1 g/kg PCE (group 3), or 1 g/kg PCE (group 4). Group 1 (negative control) received saline injections and 10% DMSO; group 5 (PCE control) received saline injections and 1 g/kg PCE. Blood was collected at weeks 3, 5, and 10 for serum cytokine measurement (IL-6, IL-1β, TNF-α) by ELISA. At week 14, fecal samples were collected for 16S rRNA gene sequencing (V4 region, Illumina NovaSeq 6000) to analyze gut microbiota. At week 15, rats were sacrificed and colons were stained with methylene blue for ACF enumeration. In a separate short-term inflammation study, colonic mucosal mRNA expression of IL-6, IL-1β, and TNF-α was measured by qPCR. In vitro, the effect of PCE on IL-6-induced proliferation of HCT-116 and HT-29 colon cancer cells was assessed by MTT assay, and cytokine secretion from LPS-activated RAW 264.7 macrophages was measured by ELISA.
**Key Results:** PCE contained 49.65 ± 3.56 mg gallic acid equivalents/g of total phenolics and 41.02 ± 2.68 mg catechin equivalents/g of flavonoids. HPLC analysis identified 21.06 ± 2.13 mg rosmarinic acid, 11.11 ± 0.40 mg luteolin, and 6.55 ± 1.78 mg apigenin per gram of PCE. In the ACF progression study, the DMH + DSS positive control group had a mean of 388 ± 90.11 ACF/rat. High-dose PCE (1 g/kg) significantly reduced total ACF by 66.46% (130 ± 31.78 ACF/rat, p < 0.01), with the strongest inhibition in the proximal colon (77.07% reduction, p < 0.01). Low-dose PCE (0.1 g/kg) reduced ACF by 35.20% (251 ± 56.53 ACF/rat, p < 0.01). The number of crypts per focus (AC/f) was not significantly reduced. Serum cytokine levels at week 10 were significantly lower in both PCE-treated groups compared to the positive control (p < 0.05). Colonic mRNA expression of IL-6 and TNF-α was significantly decreased in the high-dose PCE group (1.23 ± 0.00 and 1.29 ± 0.08-fold, respectively, p < 0.05), while IL-1β mRNA was significantly reduced in both PCE dose groups (1.28 ± 0.07 and 1.08 ± 0.02-fold, p < 0.05). Gut microbiota analysis showed that DMH + DSS treatment reduced alpha diversity (Shannon and Simpson indices, p < 0.05), which was restored by PCE in a dose-dependent manner. The relative abundance of Gammaproteobacteria (including Escherichia/Shigella) was increased in DMH + DSS rats but reduced by PCE. PCE also increased beneficial bacteria such as Muribaculaceae, Lactobacillus, and Oscillospiraceae. In vitro, PCE at 100 and 200 µg/mL significantly inhibited IL-6-induced proliferation of HCT-116 (by ~12% and ~25%, p < 0.05 and p < 0.01) and HT-29 cells (by ~12% and ~24%, p < 0.05 and p < 0.01) after 48 h. In LPS-activated RAW 264.7 macrophages, PCE (100 and 200 µg/mL) significantly reduced secretion of IL-6, IL-1β, and TNF-α (p < 0.01).
**Clinical Implications:** This study provides the first evidence that Perilla frutescens seed residue extract can suppress inflammation-induced colon carcinogenesis in a rat model by reducing systemic and local inflammation, modulating gut microbiota toward a healthier profile, and inhibiting cancer cell proliferation under inflammatory conditions. The human equivalent dose was estimated at 9.73 g/day of PCE (or 132.5 g of PF seed residue) for a 60 kg adult, providing approximately 204.9 mg rosmarinic acid, 108.10 mg luteolin, and 63.73 mg apigenin daily. While these findings are promising, the study is limited by its animal model design, and the specific mechanisms linking PCE-mediated microbiota changes to inflammation and cancer progression require further investigation. Nonetheless, PF seed residue—a low-value byproduct of oil production—represents a potentially cost-effective source of bioactive compounds for colon cancer chemoprevention.