**Background:** Sweet potato (Ipomoea batatas) leaves are highly nutritious but often underutilized as a vegetable. Light quality, particularly red and blue wavelengths from LEDs, is known to influence plant growth and metabolite accumulation. This study aimed to investigate the effects of monochromatic red and blue LEDs on the physiological, biochemical, metabolomic, and transcriptomic profiles of sweet potato leaves to understand how light quality can be used to improve leaf quality for human nutrition.
**Methods:** Sweet potato vines ('Fushu-18') were grown in a controlled chamber under either 100% red (632 nm) or 100% blue (462 nm) LEDs at a low photosynthetic photon flux density of 50 μmol m⁻² s⁻¹, with a 16/8 h photoperiod, 25±2°C, and 70±5% relative humidity. After three weeks of exposure, leaves were harvested for analysis. Physiological parameters measured included membrane stability index (MSI), leaf water content (LWC), leaf area, fresh and dry weights, number of abscised leaves, and intumescence percentage. Biochemical analyses included chlorophyll a and b, total chlorophyll, carotenoids, total soluble sugar, total soluble protein, vitamin C, total phenolic content (TPC), total flavonoid content (TFC), and DPPH antioxidant scavenging activity. Metabolome profiling was performed using UPLC-ESI-MS/MS, and transcriptome analysis (RNA-Seq) was conducted on an Illumina platform. Differentially accumulated metabolites (DAMs) were identified with thresholds of |log₂FC| ≥ 1 and VIP ≥ 1. Differentially expressed genes (DEGs) were defined by FDR < 0.01 and fold change ≥ 2. KEGG pathway enrichment analyses were performed for both DAMs and DEGs. qRT-PCR validated nine randomly selected genes.
**Key Results:** Blue light significantly improved several physiological parameters: membrane stability (93±0.33% vs 90±1.06%), leaf fresh weight (0.50±0.04g vs 0.32±0.01g), leaf dry weight (0.05±0.01g vs 0.02±0.00g), and reduced leaf abscission (25±2.4% vs 47±2.7%) and intumescence (13±1.6% vs 25±1.8%) compared to red light. However, red light resulted in higher chlorophyll a (18.41±0.66 vs 16.47±0.60 mg g⁻¹), chlorophyll b (7.26±0.37 vs 5.33±0.29 mg g⁻¹), total chlorophyll (25.67±1.03 vs 21.80±0.89 mg g⁻¹), carotenoids (4.07±0.15 vs 3.46±0.12 mg g⁻¹), total soluble sugar (0.7±0.02% vs 0.5±0.02%), and vitamin C (19.83±0.77 vs 10.4±0.38 mg g⁻¹). Conversely, blue light increased total soluble protein (1.3±0.09% vs 0.7±0.01%), total phenolics (259.3±8.61 vs 199.8±1.61 mg g⁻¹), total flavonoids (54.7±0.95 vs 40.1±0.64 mg g⁻¹), and DPPH antioxidant activity (94.9±0.14% vs 91.1±0.58%). Metabolome analysis detected 744 compounds. A total of 95 DAMs were identified: 77 were upregulated under red light and 18 under blue light. Red light increased 23 amino acids and derivatives (including eight essential amino acids), 25 free fatty acids (including stearidonic acid), and nine organic acids. Blue light increased seven phenolic acids (including caffeic acid) and four vitamin B6 compounds. KEGG enrichment showed that linoleic acid metabolism (ko00591) and alpha-linolenic acid metabolism (ko00592) were the most significantly enriched pathways. Transcriptome analysis identified 17,995 genes, with 14,039 novel genes annotated. A total of 615 DEGs were found: 510 downregulated and 105 upregulated under red light compared to blue light. Blue light upregulated three anthocyanin biosynthetic genes (encoding anthocyanidin 3-O-glucosyltransferase) and seven carotenoid biosynthetic genes (encoding carotenoid isomerase, beta-carotene 9-cis-all-trans isomerase, zeaxanthin, antheraxanthin, and abscisate enzymes).
**Clinical Implications:** This study demonstrates that manipulating LED light quality can significantly alter the nutritional and biochemical profile of sweet potato leaves. Blue light enhances antioxidant compounds (phenolics, flavonoids) and protein content, while red light boosts essential amino acids, fatty acids, and sugars. These findings provide a basis for optimizing controlled-environment production of sweet potato leaves as a nutrient-dense vegetable, potentially improving human nutrition, especially in regions where sweet potato is a staple crop. The identified genes and metabolites could be targeted in breeding programs to enhance leaf quality.