Newly discovered coffee compounds beat diabetes drug in lab tests

Three newly identified compounds were found to strongly inhibit α-glucosidase, an enzyme that plays a central role in breaking down carbohydrates during digestion. Because this enzyme directly affects how quickly sugars enter the bloodstream, the discovery points to possible new functional food ingredients aimed at managing type 2 diabetes.

Functional foods offer more than basic nutrition. Many contain naturally occurring molecules that may support health, including compounds with antioxidant, neuroprotective, or glucose-lowering effects. Finding these helpful substances is difficult because foods are chemically complex. Older discovery methods can be slow and inefficient, which has pushed researchers to adopt more advanced tools such as nuclear magnetic resonance (NMR) and liquid chromatography-mass spectrometry (LC-MS/MS). These techniques are especially valuable for studying roasted coffee, which contains a wide range of overlapping chemical components.

Study Reveals Anti Diabetic Potential in Coffee

Researchers led by Minghua Qiu at the Kunming Institute of Botany, Chinese Academy of Sciences, reported their findings in Beverage Plant Research. Their work highlights previously unknown anti-diabetic activity in coffee and adds new insight into its role as a functional food.

The team designed a three-step, activity-focused process to uncover bioactive diterpene esters in roasted Coffea arabica beans. Their approach aimed to detect both common and extremely low-level compounds that could inhibit α-glucosidase, while also reducing solvent use and speeding up analysis.

First, the crude diterpene extract was separated into 19 fractions using silica gel chromatography. Each fraction was then analyzed with ^1H NMR and tested for α-glucosidase inhibition. By applying cluster heatmap analysis to the ^1H NMR data, the researchers identified Fr.9-Fr.13 as the most biologically active fractions based on distinctive proton signal patterns.

Further analysis of a representative sample, Fr.9, using ^13C-DEPT NMR revealed the presence of an aldehyde group, confirming earlier findings. After purification with semi-preparative HPLC, the scientists isolated three previously unknown diterpene esters, named caffaldehydes A, B, and C. Their chemical structures were verified through 1D and 2D NMR along with high-resolution mass spectrometry (HRESIMS).

Stronger Effects Than a Common Diabetes Drug

Although the three caffaldehydes differed in their fatty acid components (palmitic, stearic, and arachidic acids), all showed notable α-glucosidase inhibition. Their IC₅₀ values were 45.07, 24.40, and 17.50 μM respectively, indicating stronger activity than the comparison drug acarbose.

To uncover additional trace compounds that were difficult to detect using NMR or HPLC alone, the team applied LC-MS/MS to combined fraction groups. They then built a molecular network using GNPS and Cytoscape. This analysis revealed three more previously unknown diterpene esters (compounds 4-6) that were closely related to caffaldehydes A-C. While they shared similar fragment patterns, these molecules contained different fatty acids (magaric, octadecenoic, and nonadecanoic acids). Searches of existing compound databases confirmed that these substances had not been reported before.

Together, the results show that this integrated dereplication strategy is highly effective for identifying structurally diverse and biologically meaningful compounds in complex foods such as roasted coffee.

What This Means for Functional Foods and Future Research

The findings suggest new opportunities to develop coffee-based functional foods or nutraceuticals that support glucose control and may help manage diabetes. Beyond coffee, the same low-solvent, high-precision screening approach could be applied to other complex food sources to rapidly uncover health-related compounds. Future studies will focus on testing the biological effects of the newly discovered trace diterpenes and evaluating their safety and effectiveness in vivo.

Reference:

Guilin Hu, Chenxi Quan, Abdulbaset Al-Romaima, Haopeng Dai, Minghua Qiu. Bioactive oriented discovery of diterpenoids in Coffea arabica basing on 1D NMR and LC-MS/MS molecular network. Beverage Plant Research, 2025; 5 (1): 0 DOI: 10.48130/bpr-0024-0035