A common FUT2 variant determines whether ABO blood-group antigens appear in saliva, gut mucus and other secretions; the roughly one in five people who are "non-secretors" are strongly resistant to the dominant norovirus strains.
Everyone has an ABO blood type, but not everyone displays it in the same places. Most people, called secretors, also show their ABO antigens dissolved in saliva, tears, gut mucus, cervical fluid and other bodily secretions, in addition to on their red blood cells. A minority of people, called non-secretors, have completely normal ABO blood types but do not display those antigens in their secretions at all. Whether someone is a secretor or non-secretor is controlled almost entirely by a single gene, FUT2, which makes the enzyme that builds the chemical scaffold those antigens are attached to outside the bloodstream.
Secretor status is inherited straightforwardly: one working copy of FUT2 is enough to be a secretor, so non-secretor status only appears in people who inherit a non-working copy from both parents. Roughly one in five people of European ancestry are non-secretors, and similar frequencies are seen in some African populations, though the specific gene change responsible differs somewhat by ancestry. It is a completely normal, common trait, not a disease, and most non-secretors go through life with no idea they are one — but it turns out to have some genuinely interesting and well-documented downstream effects.
Norovirus, the leading cause of acute gastroenteritis outbreaks worldwide, needs to attach to a specific carbohydrate structure on gut cells to establish infection, and that structure is built using the same FUT2 enzyme that creates secreted ABO antigens. Non-secretors largely lack this attachment point in their gut lining. As a result, people who are genetically non-secretors are strongly resistant to symptomatic infection with the GII.4 norovirus strains that cause the majority of outbreaks globally — volunteer challenge studies and outbreak investigations have repeatedly found this. The protection is not absolute (occasional breakthrough infections, including with other norovirus strains that use different attachment routes, have been documented), but it is one of the most reproducible gene-infection relationships in human genetics.
Large genetic studies have consistently found that FUT2 genotype is one of the strongest common genetic influences on measured blood vitamin B12 levels, with non-secretors tending to have somewhat lower measured levels. Part of this reflects a real biological difference in how B12 is bound and carried in blood, but part of it may also be a quirk of how the standard blood test measures B12 rather than a true difference in the amount of vitamin available to tissues — a distinction worth knowing if you ever see this variant mentioned alongside a borderline B12 level.
Because secretor status changes the sugar molecules lining the gut, it also shapes which bacteria can attach and thrive there. Some studies have found that non-secretors, particularly breastfed infants, host different, often lower amounts of certain beneficial gut bacteria such as Bifidobacterium. This is a genuinely interesting research finding, but larger studies looking broadly across the gut microbiome in adults have found the effect to be inconsistent, so it should be treated as a real but not fully settled piece of the picture rather than a dependable, universal rule.
Separately, large genetic studies spanning tens of thousands of people have found that non-secretor status is associated with a modestly increased risk of both Crohn's disease and type 1 diabetes, and these associations have been confirmed across multiple independent studies at a level of statistical certainty that is difficult to dismiss as chance. It's important to be precise about scale, though: these are population-level shifts in relative risk, not something that turns a healthy non-secretor's odds of developing either disease into anything close to a coin flip. The great majority of non-secretors never develop either condition.
Secretor status is a fascinating example of how one gene can sit at the crossroads of infection resistance, nutrient metabolism, the microbiome and autoimmune risk all at once. But it is not a diagnostic test for any disease, and it is not currently used to guide medical decisions in clinical practice. Knowing your secretor status is interesting biology, not a health verdict.
FUT2 (19q13.33) encodes galactoside 2-alpha-L-fucosyltransferase 2, an enzyme that adds a fucose residue to precursor glycans in secretory epithelia (salivary gland, gastrointestinal and genitourinary mucosa), generating the H-antigen scaffold onto which ABO blood-group antigens are subsequently built in secretions. A functional allele (Se) is dominant; the secretor phenotype requires only one working copy. The non-secretor phenotype requires homozygosity for loss-of-function alleles.
The variant catalogued as rs601338 is a nonsense substitution, described on the current MANE Select transcript as NM_000511.6:c.461G>A, NP_000502.4:p.Trp154Ter. Older literature numbers the identical nucleotide change against a different transcript reference and reports it as 428G>A, p.Trp143Ter (also written W143X); both designations refer to the same single-base substitution creating a premature stop codon that truncates the enzyme before it can fold into an active form. This is the predominant inactivating allele in European and African populations. In East and Southeast Asian populations, non-secretor status is more often caused by a distinct missense variant, rs1047781 (A385T, p.Ile129Phe), which also inactivates the enzyme; rs601338 genotype alone is therefore a less complete predictor of secretor phenotype in those ancestries.
Homozygosity for inactivating FUT2 alleles (non-secretor status) is found in approximately 20% of individuals of European ancestry and at broadly comparable frequencies in some African populations; frequencies and the specific causal allele differ by ancestry as above. As with any recessive trait, the ~20% figure describes the homozygous (non-secretor) genotype frequency directly; it should not be treated as the frequency of any single allele.
The dominant pandemic GII.4 norovirus genotype binds histo-blood group antigen structures displayed on secretor intestinal epithelium as an attachment factor for infection. Thorven et al. (2005) reported that individuals homozygous for the 428G>A (rs601338) nonsense allele were resistant to symptomatic infection during defined norovirus outbreaks; subsequent volunteer challenge and outbreak studies have replicated strong, though not absolute, protection against symptomatic GII.4 disease in non-secretors, with rare documented breakthrough infections and with resistance being less complete against some non-GII.4 genogroups. This remains among the most robust and mechanistically well-understood gene-infection associations described in humans.
Genome-wide association studies (Hazra et al., 2008) identified FUT2 variants in strong linkage disequilibrium with rs601338 as among the strongest common genetic determinants of circulating vitamin B12 concentration, with non-secretor genotype associated with lower measured levels. Subsequent mechanistic work indicates this reflects, at least in part, secretor-status-dependent glycosylation of haptocorrin (a B12-binding transport protein) affecting the holo-haptocorrin fraction measured by standard assays, rather than solely a difference in true tissue B12 sufficiency; holo-transcobalamin, the fraction more directly relevant to cellular B12 delivery, is less strongly affected. Clinically, this means FUT2 genotype is a documented confounder of interpreting a low-normal or borderline serum B12 result and does not by itself establish true B12 deficiency.
Early targeted studies (e.g., Wacklin et al., 2011) reported reduced abundance and diversity of Bifidobacterium species in non-secretor adults and altered Bifidobacterium colonization patterns in breastfed non-secretor infants, proposed to reflect loss of fucosylated glycan attachment sites and nutrient sources used by these taxa. However, larger population-based microbiome studies and genome-wide microbiome association analyses have found this relationship to be inconsistent or attenuated in general adult cohorts. The evidence should be characterized as moderate and context-dependent (strongest in early infancy) rather than a uniformly replicated adult finding.
FUT2 non-secretor status is a confirmed, genome-wide-significant susceptibility locus for Crohn disease, first reported by McGovern et al. (2010) and subsequently corroborated in large multi-cohort inflammatory bowel disease meta-analyses (e.g., Jostins et al., 2012, PMID 23128233), with modest per-genotype effect sizes typical of common inflammatory bowel disease risk loci. For type 1 diabetes, Smyth et al. (2011), analyzing over 8,000 cases and 10,000 controls plus family-based replication, reported a robust, genome-wide-significant association between the non-secretor (AA) genotype and increased risk (odds ratio approximately 1.2-1.3 relative to secretor genotypes). Both associations are well-replicated at the level of statistical genetics but correspond to modest shifts in absolute individual risk, consistent with a role for altered mucosal microbe-host interaction as one contributing factor among many in these polygenic, multifactorial diseases.
Genotyping rs601338 (and, where relevant, rs1047781) indicates predicted secretor/non-secretor status and the associated norovirus GII.4 resistance and altered measured B12 handling described above. It does not diagnose or exclude Crohn disease, type 1 diabetes, or any microbiome-related condition, does not eliminate the small residual risk of norovirus infection, and is not used as a basis for clinical management of any of the above conditions.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about FUT2 Secretor Status comes down to these specific, well-studied positions — not a diagnosis.
It means your ABO blood-group antigens are present on your red blood cells as usual, but are not displayed in your saliva, gut mucus, and other secretions the way they are in most people. It's a completely normal, common genetic trait carried by roughly one in five people of European ancestry, not a disorder.
It means you are strongly resistant to symptomatic infection from the GII.4 norovirus strains that cause most outbreaks, because those strains need an attachment molecule that non-secretors largely lack in their gut lining. It is not absolute protection — occasional breakthrough infections and infections with other norovirus strains have been documented — so normal food and hand hygiene still matters.
Not necessarily. Non-secretor genotype is strongly associated with lower measured blood B12 levels, but part of that reflects how the standard lab test measures a B12-carrying protein rather than a true shortage of B12 available to your cells. A borderline B12 result in a non-secretor may need clinical interpretation rather than being read as straightforward deficiency.
No. Large studies have found that non-secretor status is associated with a modestly higher relative risk of both conditions, and the statistical evidence for the association itself is strong. But the absolute effect on any one person is small, both conditions are caused by many genes plus environmental factors together, and the large majority of non-secretors never develop either one.
It is real but less consistent than the norovirus or vitamin B12 findings. Some studies, especially in breastfed infants, have found clear differences in beneficial gut bacteria like Bifidobacterium between secretors and non-secretors. Larger studies in general adult populations have found this effect to be weaker or inconsistent, so it should be treated as a promising but unsettled area rather than a firm rule.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.