Professional · Article 24
Nutrition & Grain Science
The nutrition of bread comes down to three physical facts: which fractions of the kernel survive milling, how finely the grain is ground, and how the dough is fermented. Get those right and you have done more than any additive can.
Health claims here are stated as evidence, not advocacy, and every contested claim carries an explicit grade. The consolidated ledger at the foot of the article is the honest summary.
The kernel, and what milling takes
A cereal grain is botanically a caryopsis, a single-seeded dry fruit whose seed coat is fused to the fruit wall. Every grain is organised into three tissues, and baking nutrition starts with knowing what each carries.
The bran is a multilayered outer shell: pericarp, testa, and the aleurone layer, which is botanically the outermost cell layer of the endosperm but leaves with the bran during milling. It concentrates insoluble fibre, the majority of the kernel’s B-vitamins, its minerals, and its phytochemicals. Wheat bran is one of the richest dietary sources of ferulic acid. The aleurone also holds most of the kernel’s phytic acid and the enzyme phytase, a pairing that becomes the whole fermentation story below.
The germ is the embryo, the living part that would become a plant. Small in mass but dense in nutrition, it carries polyunsaturated oils, vitamin E, folate, additional B-vitamins, and high-quality protein. Wheat germ is roughly 10% oil, largely linoleic acid, which is the source of both its value and its instability.
The endosperm is the seed’s starch-and-protein food store, overwhelmingly starch plus the gliadins and glutenins that hydrate into gluten. It is comparatively poor in fibre, vitamins, minerals, and phytochemicals, because those live in the two fractions milling removes.
The implication is direct: the most nutrient-dense parts of the grain are exactly the parts refining takes out, and what remains is mostly digestible starch.
Extraction rate is the master variable
Extraction rate is the percentage of the grain, by weight, that ends up in the flour. A 100% extraction flour is whole grain. Typical white bread flour runs roughly 72 to 75% extraction, meaning the 25 to 28% removed is the bran and germ. High-extraction or half-white flours, often 80 to 90%, sit in between, keeping more nutrition at some cost to lift and crumb colour.
Roller milling passes grain through successive pairs of corrugated then smooth steel rollers, sifting between stages. Its defining capability is separation: it cleanly shears endosperm away from bran and germ, letting the miller produce pure white flour and divert the germ oil out of the flour stream entirely. That is a commercial triumph, white flour of consistent quality with long shelf life, and a nutritional subtraction.
Stone milling crushes the whole kernel between two stones. It does not cleanly separate anything, so the germ and its oil are crushed into the flour. Whole stone-ground flour therefore keeps the germ, its oil, vitamin E, and folate, and generally carries more flavour.
The marketing claim that stone-ground flour is categorically more nutritious is only partly right. It is true chiefly when the flour is whole, because the germ is retained. A bolted, sifted stone flour that removes bran and germ is not nutritionally superior to roller-milled white flour. The real variable is what fraction is kept, not the grinding mechanism.
Stone milling also generates more frictional heat, which can degrade heat-sensitive nutrients and accelerate oil oxidation, and because germ oil is now dispersed throughout, stone-ground whole flour goes rancid faster.
What “whole grain” actually means
The operative definition requires that bran, germ, and endosperm be present in the same relative proportions as they exist in the intact kernel. Processing is allowed. A grain can be cracked, rolled, extruded, cooked, or milled into flour, provided the three fractions survive together in their original ratio.
That reconstitution rule is what lets whole-wheat flour count despite being pulverised, and it is what makes label reading hard. “Multigrain,” “stone-ground,” “wheat flour,” and “made with whole grains” do not guarantee whole grain, and brown colouring is frequently caramel colour or molasses. The reliable signals are “100% whole grain” and a whole grain as the first ingredient.
The evidence on whole versus refined
The epidemiological case is among the more consistent in nutrition science. The landmark dose-response meta-analysis of 45 prospective studies found that each additional 90 g per day of whole grains, about three servings, was associated with the following reductions in risk.
Later work commissioned by the WHO reinforced that higher intakes of fibre and whole grains track with lower all-cause and cardiovascular mortality, type 2 diabetes, and colorectal cancer, in a graded fashion. For type 2 diabetes specifically, cohorts consistently associate higher whole-grain intake with roughly 20 to 30% lower incidence comparing high to low consumers.
This evidence is overwhelmingly observational. Whole-grain eaters tend to be more health-conscious in general, which is residual confounding, and randomised trials, which mostly measure intermediate markers over weeks, show more modest and sometimes mixed effects than the cohort associations imply.
The consensus judgment: the association is real and partly causal, plausible through fibre, lower glycemic response, minerals, and phytochemicals. But the magnitude from cohort studies likely overstates the pure grain effect. Association strong; precise causal magnitude moderate.
Four mechanisms plausibly underlie the benefit. Fibre, both insoluble for bulk and transit, and soluble and fermentable, arabinoxylan and beta-glucan, feeding gut microbiota to yield short-chain fatty acids linked to insulin sensitivity and satiety. Micronutrients, particularly magnesium, itself inversely associated with diabetes, plus zinc, selenium, and B-vitamins. Phytochemicals, including ferulic acid and other phenolics, lignans, alkylresorcinols, and phytosterols. And glycemic response, where physical structure turns out to matter more than the label.
Why particle size decides the glycemic story
Intact and coarsely processed whole grains blunt the postprandial glucose and insulin spike relative to finely milled refined starch. Whole kernels and coarse grinds, stoneground, cracked, steel-cut, digest slowly.
Finely milled whole-wheat flour, however, has a glycemic index nearly as high as white flour, because grinding destroys the intact cell walls that slow starch access. White bread sits around GI 70 to 75, and many finely milled whole-wheat breads are similar.
Whole grain delivers its full glycemic benefit only when particle size stays large. It is the intactness, not the whole-grain label, that lowers glycemic index. This is one of the most persistent misconceptions in health messaging about bread.
A second one worth flagging: fat and sugar-rich baked goods can show a deceptively moderate glycemic index because fat slows digestion, even though they are calorically dense. Glycemic index is not a proxy for healthy.
Three levers lower it: intact or coarse structure, soluble viscous fibre such as beta-glucan from oats and barley, and acidity from sourdough.
The baking tension
Whole-grain baking is genuinely harder, for reasons rooted in the same fractions that carry the nutrition.
Bran physically disrupts gluten. Sharp bran particles act like tiny blades, cutting the developing network and puncturing gas cells, which reduces loaf volume and gives a denser crumb. Bran and fibre compete for water, since arabinoxylans absorb a great deal, so whole-grain doughs need higher hydration and an under-hydrated one is stiff and tears. The germ oil and enzymes bring rancidity risk and enzymatic activity, lipoxygenase and proteases, that can weaken dough. And bran’s tannins and ferulic acid can further interfere with rheology.
Bakers mitigate with soaking or autolyse, hydrating the bran before mixing to soften its edges, higher hydration, longer and gentler mixing, added vital wheat gluten, and fermentation, which softens bran and improves texture and nutrition at the same time. The tension is real: maximum nutrition and maximum lightness pull in opposite directions, and most artisan practice is a negotiated middle.
What fermentation actually does
Phytate degradation is the strongest claim, and it is genuinely strong. Whole grains are rich in phytic acid, which chelates iron, zinc, magnesium, and calcium in the gut and inhibits their absorption. This is why whole-grain diets, despite higher mineral content, can deliver lower mineral availability. Grains also carry phytase, in the aleurone, and phytase has a pH optimum in the acidic range, roughly 4.5 to 5.5.
Sourdough fermentation drops dough pH through lactic and acetic acid production, which drives endogenous phytase into its active range. Reported phytate degradation in optimised sourdough reaches roughly 60 to 90% or more, against about 20 to 40% in straight yeast dough, with measured improvements in iron, zinc, and magnesium bioaccessibility. This is the one place where whole-grain sourdough is genuinely nutritionally superior to both white bread and yeasted whole-grain bread.
Fructan reduction is well supported and clinically relevant. Wheat contains fructans, a fermentable oligosaccharide and a leading trigger of gas, bloating, and pain in irritable bowel syndrome. Sourdough bacteria and yeast consume fructans during fermentation, and sufficiently long fermentation can lower a bread’s FODMAP load below the low-FODMAP threshold. This is a plausible, partly demonstrated explanation for why many people who believe they react to wheat tolerate genuine long-fermented sourdough: their trigger may be fructans rather than gluten. The caveat matters commercially, since many breads sold as sourdough are short-fermented or use added yeast and remain high in fructans.
Glycemic response is moderately supported. Sourdough’s organic acids lower the postprandial glucose and insulin response relative to yeasted bread of the same flour, probably by slowing gastric emptying, reducing starch digestibility, and possibly inhibiting amylase. Controlled feeding studies generally show a meaningfully lower glycemic index, though magnitude varies and not every trial is positive.
It is not probiotic. Baking inactivates the microbes. The benefits are prebiotic and fermentation-derived, not live culture.
It is not celiac-safe. Long fermentation partially breaks down gluten, which is real, but not to a degree that makes ordinary sourdough safe for people with celiac disease. The gluten-free threshold remains below 20 parts per million, and regular sourdough does not reach it.
Acrylamide
Acrylamide is a process contaminant formed above about 250 °F (120 °C) as a branch of the Maillard reaction, from the amino acid asparagine reacting with reducing sugars. It forms in bread crust, cookies, crackers, and toast, and the darker the crust, the more of it there is.
It is a rodent carcinogen and a neurotoxin at high doses, classified by IARC as probably carcinogenic to humans, Group 2A. Human dietary epidemiology is inconsistent and has not clearly demonstrated cancer risk at dietary exposures, but regulators advise keeping intake as low as reasonably achievable.
Mitigation in baking is straightforward: bake to golden rather than dark brown, reduce time or temperature where you can, prefer lower-reducing-sugar formulations, and avoid over-proofing, which raises free asparagine. Industrially, the enzyme asparaginase strips asparagine before baking. The chemistry is well characterised; the human dietary risk is uncertain, and the appropriate framing is precautionary rather than alarmist.
The evidence ledger
A consolidated, honest accounting, which is what a reference owes its readers.
The throughline: the nutrition of baked goods is governed by three physical facts. Which fractions of the kernel survive milling, since bran and germ carry both the nutrition and the spoilage. How finely the grain is ground, since intactness governs glycemic response. And how the dough is fermented, since acid and enzymes govern mineral availability, fructans, and glycemia. Get those three right and you have improved the nutrition of bread more than any additive can.
References
- Aune, D., Keum, N., Giovannucci, E., et al. (2016). Whole grain consumption and risk of cardiovascular disease, cancer, and all cause and cause specific mortality: Systematic review and dose-response meta-analysis of prospective studies. BMJ, 353, i2716.
- Reynolds, A., Mann, J., Cummings, J., et al. (2019). Carbohydrate quality and human health: A series of systematic reviews and meta-analyses. The Lancet, 393(10170), 434–445.
- Leenhardt, F., Levrat-Verny, M.-A., Chanliaud, E., & Rémésy, C. (2005). Moderate decrease of pH by sourdough fermentation is sufficient to reduce phytate content of whole wheat flour through endogenous phytase activity. Journal of Agricultural and Food Chemistry, 53(1), 98–102.
- Lopez, H. W., Krespine, V., Guy, C., et al. (2001). Prolonged fermentation of whole wheat sourdough reduces phytate level and increases soluble magnesium. Journal of Agricultural and Food Chemistry, 49(5), 2657–2662.
- Laatikainen, R., Koskenpato, J., Hongisto, S.-M., et al. (2017). Pilot study: Comparison of sourdough wheat bread and yeast-fermented wheat bread in individuals with wheat sensitivity and irritable bowel syndrome. Nutrients, 9(11), 1215.
- Kasarda, D. D. (2013). Can an increase in celiac disease be attributed to an increase in the gluten content of wheat as a consequence of wheat breeding? Journal of Agricultural and Food Chemistry, 61(6), 1155–1159.
- Monash University. FODMAP and sourdough bread. https://www.monashfodmap.com
- Whole Grains Council. (2004). Definition of a whole grain. https://wholegrainscouncil.org
- Cereals & Grains Association (formerly AACC International). Whole grain definition. https://www.cerealsgrains.org
- U.S. Food and Drug Administration. (2016). Guidance for industry: Acrylamide in foods.
- European Food Safety Authority. Acrylamide in food. https://www.efsa.europa.eu
- International Agency for Research on Cancer. Acrylamide (Group 2A monograph).
- Centers for Disease Control and Prevention. Folic acid fortification and neural tube defects. https://www.cdc.gov
- McGee, H. (2004). On Food and Cooking: The Science and Lore of the Kitchen (rev. ed.). Scribner.
- Myhrvold, N., & Migoya, F. (2017). Modernist Bread. The Cooking Lab.
- Gänzle, M. G. Sourdough microbiology and metabolism reviews. Food Microbiology and related reviews.
Print references: Modernist Bread (Myhrvold & Migoya, 2017); McGee, On Food and Cooking (2004); Aune et al., BMJ (2016); Reynolds et al., The Lancet (2019).