Why the same carbohydrate intake produces different metabolic outcomes in different people.
The debate over carbohydrates in diet has produced more conflicting advice than almost any other nutritional topic. Low-carb diets produce dramatic results for some people and modest results for others. High-carbohydrate diets are associated with optimal health in some populations and with metabolic problems in others. Individual variation in carbohydrate metabolism is a significant part of the explanation for these apparent contradictions, and genetics is a significant driver of that variation.
AMY1 encodes salivary amylase, the enzyme in saliva that begins the digestion of starch in the mouth. Uniquely among nutritionally relevant genes, AMY1 varies not just in sequence but in copy number. Different people carry different numbers of copies of the AMY1 gene, ranging from two to more than fifteen.
People with higher AMY1 copy numbers produce more salivary amylase and digest starch more efficiently. Their blood glucose response to starchy foods is lower and more gradual. People with fewer AMY1 copies produce less amylase, digest starch more slowly, and have a higher and faster blood glucose response to the same starchy meal.
Research has found that low AMY1 copy number is associated with higher obesity risk, and that this association is strongest in the context of high starch intake. For people with low copy numbers, starchy foods produce a substantially greater metabolic burden than they do for high-copy-number individuals eating the same meal. This is a source of individual variation in carbohydrate response that cannot be corrected by willpower or awareness, only by adjusting intake.
TCF7L2 is the gene most strongly associated with type 2 diabetes risk in genome-wide association studies. It encodes a transcription factor involved in insulin secretion from pancreatic beta cells. Variants in TCF7L2, particularly rs7903146, are associated with reduced insulin secretion in response to blood glucose elevation, meaning glucose levels after a carbohydrate-containing meal remain elevated for longer than in people without the variants.
The practical implication is not that TCF7L2 risk variant carriers are diabetic or pre-diabetic. It is that their glycaemic response to carbohydrate intake is higher than average, and that dietary approaches that moderate refined carbohydrate intake are likely to be more beneficial for them than for the general population. This is a case where genetic information provides a rational basis for a dietary choice that might otherwise seem like personal preference or fad.
PPARG influences insulin sensitivity and affects how the body partitions energy between fat storage and other uses in response to dietary intake. Covered in more detail in the fat metabolism article, PPARG is also relevant here because insulin sensitivity is the primary mechanism through which carbohydrate intake affects metabolic outcomes.
People with certain PPARG variants have lower baseline insulin sensitivity, which means carbohydrate intake has a larger effect on fat storage and blood glucose stability than it does for those with higher baseline insulin sensitivity. The relationship between carbohydrate metabolism and fat metabolism is direct through this gene.
IRS1 encodes insulin receptor substrate 1, a key signalling protein in the insulin pathway. Variants in IRS1 affect insulin signalling efficiency and are associated with insulin resistance and type 2 diabetes risk. SLC2A2 encodes GLUT2, the glucose transporter responsible for glucose uptake in the intestine, liver, and pancreatic beta cells. Variants in SLC2A2 affect glucose transport efficiency and are associated with altered carbohydrate handling.
These variants add further layers to the individual carbohydrate response picture, though with somewhat smaller individual effect sizes than TCF7L2.
The evidence from carbohydrate metabolism genetics does not support blanket low-carb advice for everyone. For people without significant risk variants in TCF7L2, AMY1, or PPARG, moderate carbohydrate intake from whole food sources is entirely consistent with good metabolic health. For people with multiple risk variants, reducing refined carbohydrate intake and choosing lower glycaemic carbohydrate sources produces measurably better metabolic outcomes than the population-average guidance alone would suggest.
The distinction between carbohydrate types matters regardless of genetics. Whole food carbohydrates with fibre, whole grains, legumes, vegetables, are absorbed more slowly and produce lower glycaemic responses than refined carbohydrates. For people with carbohydrate metabolism risk variants, this distinction matters more than it does for the average person.
The success of low-carbohydrate dietary approaches in some individuals and their underwhelming results in others is partly explained by this genetic variation. People with multiple carbohydrate metabolism risk variants experience more dramatic metabolic improvement on carbohydrate reduction because their baseline response to carbohydrate intake is higher. People without these variants experience a more modest benefit because their carbohydrate handling is closer to the average that population-level recommendations are designed for.
This does not mean that low-carb diets are only for people with genetic risk variants. But it does mean that people who find low-carb approaches disproportionately beneficial are more likely to carry these variants, and that genetic information provides a rational basis for personalising the degree of carbohydrate restriction in a way that generic dietary advice cannot.
