The motivation is rooted in a resource problem of staggering scale. Agriculture consumes nearly seventy percent of China’s total freshwater, and conventional flooded rice paddies account for roughly seventy percent of that agricultural use. Climate change, seasonal drought and competition from industry are putting this water-hungry system under increasing strain. Upland rice, by contrast, resists drought and grows on marginal land including sandy wastelands and saline-alkali soils. In Yunnan’s high-altitude environment, defined by intense solar radiation, low atmospheric pressure and seasonal water scarcity, upland cultivation has long been the dominant agricultural pattern, making the region a natural laboratory for water-saving rice.

The team designed a careful comparison. They collected the same upland rice cultivar from four distinct Yunnan regions, Honghe, Pu’er, Nujiang and Qujing, and matched it against eight major paddy varieties from Jiangxi and Heilongjiang. Every variety was sampled from three independent field plots with separate sowing, management and harvest, all gathered at the same maturity stage in 2025. The researchers then deployed a battery of analytical techniques: colorimetric amylose measurement, simulated oral, gastric and intestinal digestion to estimate a glycemic index, Rapid Viscosity Analyzer profiling of starch gelatinization, instrumental taste analysis, a trained fifteen-member sensory panel, an ultra-fast gas chromatography electronic nose, and untargeted metabolomics by ultra-high performance liquid chromatography coupled to high-resolution mass spectrometry.

The starch results were unambiguous. Yunnan upland rice carried amylose contents of 11.78 to 14.11 percent, higher than either paddy group, with Jiangxi rice lowest and Heilongjiang intermediate. Because amylose’s linear molecular chains pack into dense, ordered granules, the upland starch proved more thermally stable, gelatinizing at significantly higher temperatures, approaching 92 degrees Celsius in one sample, and resisting breakdown under heat and shear. Critically, this same structural rigidity slowed enzymatic hydrolysis during simulated digestion. The expected glycemic index of the upland samples ranged from 60.55 to 69.13, significantly lower than Jiangxi rice at 70.11 to 74.96 and Heilongjiang rice, which peaked near 79.55. The authors caution that in vitro values do not perfectly mirror human glycemic responses, but the consistent pattern points toward a slower-digesting grain.

Texture told a more nuanced story. Instrumental analysis showed the upland rice was firmer, with hardness values of 4.40 to 5.50 against 3.58 to 4.58 for Jiangxi varieties, and its grains stuck together far less, with viscosity readings of just 0.32 to 0.42 compared with up to 0.56 for paddy rice. Elasticity was equal or better, meaning the grains recovered well during chewing. However, high retrogradation values and low gel consistency revealed a weakness: the upland rice hardened noticeably on cooling, a drawback the sensory panel confirmed with lower cold-rice texture scores. Where the upland grain truly shone was flavor. Its taste values, which integrate umami, sweetness and kokumi, ranged from 8.75 to 9.01, significantly above both paddy groups, and panelists awarded it the highest odor and taste scores, with one sample reaching 21.86 points for taste.

The electronic nose explained that aromatic advantage. Across the twelve samples, the researchers tentatively identified 160 odor-active compounds, and the upland rice contained more of them than any paddy variety, with one sample yielding ninety compounds. Fruity-note molecules such as phenylacetaldehyde, hexanal, ethyl acetate and methyl butyrate appeared exclusively in the upland grain, whose hexanal abundance ran 1.2 to 1.5 times that of paddy rice. Fruity compounds made up 66.35 percent of the upland odor profile, against 51.33 and 59.05 percent for the two paddy groups, while paddy rice carried significantly more roasted notes. Combining relative odor activity values with random forest machine-learning importance scores, the team pinpointed three compounds, (Z)-2-nonenal, nonanal and 2-ethyl-3,5-dimethylpyrazine, as the chemical signatures that reliably distinguish the highland grain from its flooded counterparts.

Metabolomics deepened the picture. High-resolution mass spectrometry detected 1,240 metabolites, and supervised statistical modeling separated the upland samples into a tight, independent cluster with strong internal consistency. In total, 351 differential metabolites emerged in positive ion mode and 222 in negative mode, filtered by strict false-discovery-rate correction. K-means clustering revealed a group of sixty metabolites that accumulated preferentially in the upland rice, enriched in glycerophospholipids, stress-signaling molecules and osmoregulatory organic acids, the biochemical toolkit of a plant coping with drought. Glycerophospholipid abundance ran 1.8 to 2.1 times higher than in paddy rice, a finding the authors report for the first time in upland rice research, and these lipids correlated positively with amylose content, hinting at a mechanistic link between water stress metabolism and starch chemistry.

Amino acid metabolism emerged as the central regulatory hub. Of twenty-five differential amino acids, eighteen mapped to the amino acid biosynthesis pathway in KEGG enrichment analysis, with arginine and proline metabolism and cyanoamino acid metabolism also implicated. The upland grain displayed a high-abundance amino acid profile, with significantly elevated glycine, glutamic acid, threonine and arginine, while Heilongjiang rice sat at the low end and Jiangxi in between. This accumulation appears directly responsible for the flavor advantage: taste values and panel taste scores correlated significantly with glycine and glutamic acid, both classic flavor-active amino acids. Glycerophospholipids, meanwhile, correlated negatively with adhesiveness and breakdown, and the authors propose they may competitively bind water during gelatinization, limiting starch granule swelling and modulating cooked texture.

The authors are candid about the study’s limits. The upland samples came from a single cultivar while the controls spanned eight varieties, and the three sampling regions differ in altitude, latitude, climate and soil, so genotype, environment and cultivation method cannot be fully disentangled. Future experiments growing identical cultivars across sites under controlled water regimes will be needed to isolate each factor. Even so, the multi-genotype control design strengthens confidence that the quality divergence is real and stable. As climate pressure mounts on global water supplies, the study suggests that breeding and promoting drought-adapted upland rice could deliver grains that are not merely an ecological compromise but a distinctive product in their own right, aromatic, slow-digesting and metabolically unique, offering breeders a roadmap for rice that thrives with less water while pleasing the palate.

Subject of Research: Comparative quality and metabolomic profiling of Yunnan upland rice versus paddy rice

Article Title: Quality divergence of Yunnan upland rice versus Jiangxi and Heilongjiang paddy rice revealed by physicochemical, sensory, volatile and metabolomic profiling

Article References: Xiao, Y., Wang, Z., Yue, N., Tuo, Y., Wen, J., Guo, L., Zhao, C., He, X., & Nie, J. (2026). Quality divergence of Yunnan upland rice versus Jiangxi and Heilongjiang paddy rice revealed by physicochemical, sensory, volatile and metabolomic profiling. Food Chemistry: X, 39, Article 104539. https://doi.org/10.1016/j.fochx.2026.104539

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104539

Keywords: upland rice, paddy rice, Yunnan, glycemic index, amylose, metabolomics, volatile compounds, aroma, drought tolerance, starch gelatinization, amino acids, food quality

Salecan is not an ordinary hydrocolloid. Produced by microbial fermentation, it was approved as a novel food additive by the U.S. FDA in 2020 and by China’s National Health Commission in 2021. Structurally, it is a linear, unbranched glucan built from a repeating unit of seven beta-1,3-linked and two alpha-1,3-linked glucose residues, with purity exceeding 90 percent. That linearity matters enormously. Most familiar anionic polysaccharides, from xanthan gum to carrageenan and pectin, carry highly branched backbones and heterogeneous branch distributions, features that create steric hindrance and thermodynamic incompatibility with starch chains. Salecan’s naked backbone can align and entangle directly with starch molecules, while its uniformly distributed carboxyl groups from succinyl and pyruvyl substituents provide consistent negative charge across the entire chain. As a fermentation product, it is also immune to the seasonal and geographic constraints that complicate plant- and seaweed-derived gums.

To test whether Salecan’s behavior depends on starch structure, the research team led by Wanping Xiong and Xianggui Chen at Sichuan University paired two starches that sit at opposite ends of the botanical spectrum. Potato starch, a tuber starch, features large oval granules of 15 to 100 micrometers, negatively charged monophosphate ester groups on its molecular chains, a B-type crystalline pattern and a highly cross-linked granular architecture. Corn starch, the cereal archetype, presents small polygonal granules of 5 to 20 micrometers, a nearly neutral surface, an A-type crystalline pattern and a substantially higher amylose content of 25 to 28 percent with a lower degree of polymerization. Every major structural determinant of polysaccharide-starch interaction therefore differs between the pair, making it an ideal contrast model.

The divergent results began immediately. When Salecan was added at concentrations from 0.5 to 2 percent by starch dry weight, both starches showed delayed gelatinization, with pasting curves shifted to the right and peak times significantly prolonged, findings confirmed by differential scanning calorimetry. But the viscosity story split cleanly along charge lines. In potato starch, Salecan dramatically reduced peak, trough and final viscosity, and at 2 percent addition the consistency coefficient fell by 81.56 percent. The authors attribute this to electrostatic repulsion between the negatively charged potato starch and the anionic polysaccharide, which confines Salecan to the granule surface as a barrier that blocks water ingress and amylopectin leaching. Corn starch behaved in the opposite way: Salecan increased its peak viscosity, breakdown strength and final viscosity, with the consistency coefficient surging by 332.9 percent at 2 percent addition, because the neutral granules allowed Salecan to disperse through the paste and leached amylose supplied abundant hydroxyl groups for hydrogen bonding and chain entanglement.

Rheological measurements reinforced the split. All composite pastes displayed shear-thinning pseudoplastic behavior, but Salecan reduced the apparent viscosity and shear stress of potato starch systems while raising those of corn starch. Frequency sweep tests showed that Salecan depressed the storage and loss moduli of potato gels yet elevated the moduli of corn gels in a concentration-dependent fashion. For food engineers, this duality is a gift. High-viscosity tuber pastes used in sauces, custards and dessert gels become easier to pump and process without sacrificing final gel integrity, while cereal-based soups, gravies and bakery fillings gain viscosity and thermal stability exactly where hot filling and cooling demand them.

The second half of the study tackled retrogradation, the costlier enemy of shelf life, and here Salecan proved a unifying remedy. Storing gels at 4 degrees Celsius for up to 21 days, the researchers tracked gel hardness with texture analysis, water mobility with low-field nuclear magnetic resonance, crystallinity with X-ray diffraction, molecular order with Fourier transform infrared spectroscopy and microstructure with scanning electron microscopy. Every technique converged on the same conclusion: at 1 percent addition, Salecan significantly slowed the aging of both starches, cutting gel hardness by 33.63 percent for potato starch and 31.19 percent for corn starch relative to controls.

The water-migration data are particularly striking. During storage, bound water progressively escapes the starch matrix and becomes free water, a process that culminates in syneresis, the unappetizing pooling of liquid that ruins chilled desserts and fillings. Free water appeared in plain corn starch gel after only three days and in potato starch gel after fourteen days. At 1 percent Salecan, the free water fraction after 21 days fell from 3.75 percent to 0.54 percent in potato gels and from 15.81 percent to 6.31 percent in corn gels. X-ray diffraction showed that Salecan did not change the crystal type formed during aging but significantly lowered relative crystallinity, while infrared ratios revealed suppressed short-range molecular order and reduced double-helix formation at the same optimal dosage. Interestingly, both hardness and structural indices followed a U-shaped dose curve, first falling and then rebounding at high concentrations, because Salecan’s intense water-holding capacity ultimately concentrates amylose in the remaining solvent and promotes renewed association.

The proposed mechanism weaves these observations together. During short-term retrogradation, Salecan’s linear backbone entangles with amylose and sterically blocks chain re-association, delaying initial network formation. During long-term storage, it disrupts the hydrogen-bond reformation needed for double-helix nucleation, impedes the ordered alignment of amylopectin side chains, and physically retains water within the gel network. Scanning electron microscopy captured the microstructural fingerprints of these effects: Salecan produced smaller, more orderly pores and tighter matrix interconnection in potato gels, while in corn gels it preserved a smoother, looser honeycomb structure by resisting the dehydration shrinkage that normally crumples high-amylose networks. No new covalent bonds were detected, confirming that the interactions are dominated by hydrogen bonding and physical entanglement.

Perhaps the most surprising results came from simulated digestion. Freshly gelatinized starch with Salecan was hydrolyzed more slowly by alpha-amylase and amyloglucosidase, because the additive suppresses complete gelatinization and its dense composite network shields enzyme binding sites on starch molecules, a finding that points toward low-glycemic-index versions of instant porridges, steamed rice alternatives and gluten-free breads. Yet after seven days of retrogradation, the pattern inverted: Salecan-containing gels were digested faster than highly crystallized plain retrograded gels, precisely because Salecan had prevented the formation of enzyme-resistant crystalline domains. That inversion doubles as independent mechanistic evidence, confirming from a nutritional angle that Salecan suppresses resistant crystal formation during aging. The authors caution that their structural-dependence conclusions rest on just two contrasting starches, and that broader starch panels will be needed to validate generality. Still, the prospect of a single clean-label, fermentation-produced ingredient that simultaneously tames viscosity where it runs too high, boosts it where it falls short, prevents staling, stops syneresis and reshapes glycemic response is exactly the kind of multifunctional tool the clean-label food industry has been searching for.

Subject of Research: Modulation of potato and corn starch gelatinization, retrogradation and digestibility by the anionic polysaccharide Salecan

Article Title: Structure-dependent modulation of potato and corn starch gelatinization and retrogradation by Salecan

Article References: Xiong, W., Hu, L., Ni, C., Zhao, J., Huang, Y., Yang, X., Chen, P., & Chen, X. (2026). Structure-dependent modulation of potato and corn starch gelatinization and retrogradation by Salecan. Food Chemistry: X, 39, Article 104470. https://doi.org/10.1016/j.fochx.2026.104470

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104470

Keywords: Salecan, starch gelatinization, starch retrogradation, potato starch, corn starch, food hydrocolloids, rheology, X-ray diffraction, water migration, in vitro digestion, low glycemic index, clean label