Starch makes up roughly 60 to 70 percent of the wheat endosperm’s dry weight and is the single most important carbohydrate reserve in the grain. Its functionality is governed by a hierarchy of structural features: the ratio of the two polymers amylose and amylopectin, the molecular weight of those polymers, the branching pattern of the glucose chains, the way amylopectin double helices assemble into crystalline lamellae, and the overall morphology of the granules. Each of these levels influences gelatinization, hydration, retrogradation and other properties that determine whether a flour makes an airy loaf or a dense, gummy one. Because nitrogen controls how carbon is allocated during grain filling, researchers led by Md Najmol Hoque, Clair Keene and Shahidul Islam hypothesized that when nitrogen arrives relative to flowering could produce distinct, measurable changes across this entire structural hierarchy.

To test the idea, the team ran field trials during the 2022 and 2023 growing seasons at four locations in North Dakota and Minnesota — Carrington, Casselton, Prosper and Crookston — using a randomized complete block design with three commercially important hard red spring wheat cultivars: Faller, Frohberg and Glenn. Four nitrogen treatments were compared: an unfertilized control, 56 and 112 kilograms of nitrogen per hectare applied before planting, and a 168-kilogram-per-hectare split treatment that combined pre-plant and starter nitrogen with a post-anthesis top-dress of 33.6 kilograms per hectare during early grain filling. Grain was harvested at physiological maturity, milled in the laboratory, and the resulting flour subjected to an unusually comprehensive battery of analytical techniques spanning scanning electron microscopy, size-exclusion chromatography with multi-angle light scattering, X-ray diffraction, Fourier-transform infrared spectroscopy, proton nuclear magnetic resonance, differential scanning calorimetry and solvent retention assays.

The electron microscopy images told the first part of the story. Regardless of nitrogen treatment, the wheat endosperm displayed its characteristic bimodal starch granule population — large lenticular A-type granules and small spherical B-type granules — and nitrogen did not alter the intrinsic shape of individual granules. What changed was their surroundings. Under zero nitrogen, granules sat loosely packed and clearly separated with minimal matrix coverage, but as nitrogen supply increased, packing became progressively more compact and granules became increasingly embedded in a continuous protein matrix. The effect was most pronounced under the split treatment that included post-anthesis nitrogen, consistent with enhanced deposition of storage proteins around the granules during grain filling and a denser, more integrated endosperm microstructure.

The compositional data reinforced this picture. Total starch content declined steadily with increasing nitrogen, from 64.53 percent in the unfertilized control to 62.45 percent under the highest split treatment, presumably because greater nitrogen availability during grain filling favors protein deposition over carbon stored as starch. Yet starch damage during milling actually fell, from 5.08 percent to 4.62 percent, indicating that the denser endosperm matrix physically protected granules from fracture. Flour color shifted in parallel, becoming slightly darker and yellower as protein accumulation altered light scattering within the flour. These coordinated changes show that nitrogen reshapes not only how much starch the grain contains but also how mechanically robust that starch is when the grain meets the mill.

The most striking results concerned the starch polymers themselves. Amylose content followed a biphasic trajectory: it dropped from 26.33 percent in the control to 22.68 percent under the pre-plant 112-kilogram treatment, then partially recovered to 24.49 percent when post-anthesis nitrogen was added. Amylopectin moved in the mirror image, peaking at 77.32 percent under pre-plant nitrogen before easing back. Meanwhile, amylopectin molecular weight rose from 5.76 million daltons in the control to about 8.1 million daltons under nitrogen, while amylose molecular weight fell from 2.05 million to roughly 1.6 million daltons. In other words, pre-anthesis nitrogen pushed the starch toward an amylopectin-rich, high-molecular-weight architecture, whereas late-season nitrogen partially rebalanced the amylose-amylopectin ratio without dismantling the polymer architecture already established.

Structural analyses at the crystalline and molecular scales confirmed that these compositional shifts had real physical consequences. X-ray diffraction showed that all samples retained the characteristic A-type pattern of cereal starch, but relative crystallinity climbed from around 26 percent in unfertilized grain to as high as 33 percent under nitrogen, reflecting more ordered crystalline domains stabilized by the amylopectin-rich composition. Infrared spectroscopy tracked short-range molecular order in the granule’s outer regions and showed the same trend: the ratio associated with ordered structures rose while the amorphous indicator fell. Notably, the degree of branching, measured by nuclear magnetic resonance as the proportion of alpha-1,6 glycosidic linkages, continued to increase even after post-anthesis nitrogen — rising from roughly 2.4 to 2.8 percent in the control to 3.5 to 3.6 percent under the split treatment — whereas crystallinity and molecular order had already plateaued. Branching, it seems, remained responsive to nitrogen availability late into grain filling even after the major structural reorganization was complete.

Those structural differences translated directly into function. Differential scanning calorimetry revealed that gelatinization onset, peak and conclusion temperatures all rose with nitrogen, peaking under the pre-plant 112-kilogram treatment — peak gelatinization temperatures climbed from the low 60s Celsius in unfertilized grain to as high as 71.7 degrees — before easing back under the split treatment. Gelatinization enthalpy followed the same pattern, rising from about 10.5 to 12.3 joules per gram in the control to nearly 17 to 18 joules per gram under pre-plant nitrogen, meaning more energy was needed to disrupt the more highly ordered double-helical and crystalline domains. Hydration behaved differently: swelling power and solubility increased progressively with every increment of nitrogen and kept rising under post-anthesis application, with swelling jumping from 13.15 percent to 19.92 percent and solubility from 16.60 percent to 26.59 percent. Water solvent retention capacity of the flour rose while sodium bicarbonate retention fell, the latter consistent with the reduced starch damage. Late nitrogen, in short, primarily made starch more water-accessible rather than more thermally resistant.

Multivariate statistics tied the whole system together. Principal component analysis of the twelve cultivar-by-treatment combinations showed that the first two components captured 76.9 percent of the total variation, with the dominant axis describing a transition from less ordered to more ordered starch — crystallinity, molecular order, thermal stability and hydration traits all loading together on one side, and total starch, damaged starch and amylose molecular weight on the other. Correlation analysis confirmed that relative crystallinity and infrared-derived order tracked gelatinization temperatures and enthalpy, while swelling and solubility correlated with amylopectin content and structural order. Cultivar mattered too: Frohberg showed the highest gelatinization temperatures and hydration traits, Glenn maintained the highest crystallinity across treatments, and Faller exhibited the greatest branching response, with genotype-by-environment interactions significant for composition and hydration but not for polymer molecular weight, which proved more stable across environments.

The practical implications are considerable. Because pre-anthesis nitrogen chiefly governs composition, crystallinity and thermal behavior, while post-anthesis nitrogen preferentially enhances branching and hydration, fertilizer scheduling could become a tool for tailoring starch functionality to specific end uses — a harder-cooking, more thermally stable starch for one product, a more water-absorbing flour for another — rather than simply maximizing protein content. The authors caution that responses varied in magnitude across site-years that differed in temperature and rainfall, and that future work integrating starch biosynthetic enzyme activity and gene expression will be needed to explain the regulatory mechanisms. But the central message stands: the starch in a wheat kernel is not a fixed commodity. It is a dynamic structure that farmers can, in effect, redesign from the field — one carefully timed application of nitrogen at a time.

Subject of Research: Effects of pre- and post-anthesis nitrogen fertilization on starch structure and functionality in hard red spring wheat

Article Title: Differential regulation of starch structure-function relationships by pre- and post-anthesis nitrogen in hard red spring wheat

Article References: Hoque, M. N., Keene, C., & Islam, S. (2026). Differential regulation of starch structure-function relationships by pre- and post-anthesis nitrogen in hard red spring wheat. Food Chemistry: X, 39, Article 104525. https://doi.org/10.1016/j.fochx.2026.104525

Image Credits: AI Generated

DOI: 10.1016/j.fochx.2026.104525

Keywords: wheat, starch, nitrogen fertilization, post-anthesis nitrogen, amylose, amylopectin, crystallinity, gelatinization, hard red spring wheat, grain filling, food chemistry, flour quality

The review builds its framework from the ground up, beginning with the molecular architecture of starch itself. Starch is not one molecule but two: amylose, a mostly linear polymer of glucose units, and amylopectin, a hugely branched macromolecule whose short chains cluster into crystalline lamellae within the semi-crystalline starch granule. These two polymers are packed into concentric growth rings, alternating amorphous and crystalline shells that give the granule its hierarchical organization. The fine details matter enormously. The chain-length distribution of amylopectin branches determines how readily crystallites form and melt; the ratio of amylose to amylopectin governs swelling capacity, leaching behavior and gel firmness; and even the position of phosphate groups esterified onto the glucose chains, as in potato starch, can shift gelatinization temperatures and paste viscosity. Starch, in other words, is a family of materials whose members differ subtly but consequentially.

When starch meets heat and water, the granule undergoes gelatinization, the cascade of transitions that cooks care about most. The review traces this process through its successive stages: water penetrates the amorphous regions, the crystalline lamellae melt, the granule swells irreversibly to many times its original volume, amylose leaches out into the surrounding phase, and eventually the swollen granules rupture or collapse into a continuous paste. Each of these steps is a potential control point for texture. A granule that swells strongly and leaches abundant amylose produces viscous, cohesive pastes; one that resists swelling, held together by residual proteins and lipids on its surface, yields the loose, discrete texture prized in cooked rice. The authors emphasize the phenomenon of granule ghosts, remnant granular structures that persist even after gelatinization, whose integrity depends on amylose, protein and lipid content and which strongly influence whether a cooked product feels smooth or grainy.

But the granule does not act alone. The review devotes substantial attention to the minor components hitchhiking on starch granules, including surface proteins, channel-associated lipids and phospholipids that form inclusion complexes with amylose. Removing these components measurably alters swelling patterns, pasting profiles and even the cross-linking response of modified starches. Amylose-lipid complexes, formed during extrusion or baking, raise gelatinization temperatures, slow enzymatic digestion and change the rheology of the final product. Water availability emerges as another master variable: the same starch heated at low moisture behaves entirely differently from one in excess water, and competition for water among starch, proteins and other hydrocolloids in mixed systems can redirect the whole trajectory of texture development. Thermal history, including heating rate, holding time and cooling regime, completes the picture, since rapid cooling and slow cooling produce retrogradation patterns of different crystallite types and sizes.

Retrogradation, the slow reordering of gelatinized starch chains during storage, is the villain behind stale bread and the hero behind glassy rice noodles. The review distinguishes short-term retrogradation, driven by rapid amylose aggregation and gelation within hours, from long-term retrogradation, the slower recrystallization of amylopectin outer branches over days and weeks. Amylose chain length, amylopectin fine structure, storage temperature, moisture content and the presence of proteins, sugars and hydrocolloids all modulate this process. Endogenous rice proteins, for instance, have been shown to alter retrogradation kinetics, and the addition of xanthan gum or sucrose syrups can either retard or reshape crystallization. The practical consequence is that shelf-life texture, from bread firming to noodle springiness, is a moving target determined by an interacting web of variables rather than by any single compositional number.

A major contribution of the review is its critical audit of the measurement toolbox. Differential scanning calorimetry captures the energetics of crystal melting but, the authors caution, reflects only part of the gelatinization story; structural changes continue beyond the endotherm. X-ray diffraction quantifies crystallinity, while attenuated total reflectance Fourier transform infrared spectroscopy probes short-range molecular order, and the two together offer a fuller picture of structural order than either alone. Microscopy, from light microscopy to advanced electron and scanning probe techniques, reveals granule morphology and ghost integrity. The Rapid Visco Analyser provides fast, industry-friendly pasting curves, yet its fixed shear and concentration conditions limit extrapolation to real products. Rheology, including large-amplitude oscillatory shear, characterizes network mechanics under both small and large deformations, while texture profile analysis and acoustic fracture measurements attempt to bridge the gap to crispness and hardness as consumers perceive them. Sensory evaluation remains the final arbiter, and the review stresses that instrumental and sensory parameters correlate only imperfectly, demanding careful calibration.

The framework is then tested against real foods, and here the review delivers its most striking message. In reconstituted, starch-dominant matrices, purified starch gels, starch noodles, model pastes, starch structural properties are the most predictive of final texture. Amylose content, chain-length distributions and pasting parameters correlate reasonably well with gel firmness, noodle elasticity and paste viscosity. But as the matrix becomes more complex, predictability degrades. In intact grains such as cooked rice and quinoa, texture depends on the interplay of starch fine structure with protein bodies, cell walls and the geometry of the whole grain, so chemical composition of the raw grain often outperforms isolated starch measurements. In protein-containing cereal products such as bread and noodles, gluten networks compete with starch for water and mechanically constrain granule swelling, making texture a negotiated outcome of protein-starch-water interactions.

Plant tissues push conditionality even further. In potatoes, the review notes, cooked texture is shaped by parenchyma cell structure, cell-to-cell adhesion, pectin chemistry and the degree of cell separation during cooking, alongside starch swelling pressure. Mealy versus waxy potato textures reflect differences in starch fine structure between cultivars, but the cellular architecture of the tuber mediates how those differences are expressed. Sweetpotato fries and baked roots similarly show that isolated starch properties only partially predict sensory texture, with granule size, thermal properties and tissue structure all contributing. Processing-modified structures, from extruded snacks to pre-dried chips, add yet another layer, since shear, drying and frying create textures whose determinants are as much mechanical and thermal as compositional. The lesson is that starch properties are necessary but insufficient inputs for texture prediction outside the simplest systems.

The review concludes with a methodological prescription that reads almost like a manifesto for the field: reliable texture prediction requires matched processing conditions and complementary measurements spanning structure, function, instrumentation and sensory perception. A single DSC scan or RVA curve, however convenient, cannot carry the predictive burden that food scientists and breeders often place on it. Instead, the authors advocate multiscale characterization, from molecular chain architecture through granule organization to matrix-level rheology and, ultimately, human sensory testing, all conducted under processing conditions that mirror the intended application. For an industry racing to design plant-based foods, gluten-free products and climate-resilient crops, the message is both sobering and empowering: starch structure matters profoundly, but it speaks only in the accent of its surroundings. Learning that accent, the authors argue, is the key to engineering the textures consumers crave.

Subject of Research: The relationship between starch molecular structure and texture expression in starch-rich foods

Article Title: From starch structure to food texture: a review on structural determinants and context-dependent expression in starch-rich foods

Article References: Park, H.-J., Ye, S.-J., & Baik, M.-Y. (2026). From starch structure to food texture: a review on structural determinants and context-dependent expression in starch-rich foods. Food Science and Biotechnology. https://doi.org/10.1007/s10068-026-02318-2

Image Credits: AI Generated

DOI: 10.1007/s10068-026-02318-2

Keywords: starch structure, food texture, gelatinization, retrogradation, amylose, amylopectin, food matrix, rheology, sensory evaluation, differential scanning calorimetry, rice, potato