β-casein is a major protein component of cow’s milk, and numerous variants have been described, including the A1 and A2 types. The A1 and A2 types differ in terms of the amino acid at position 67, being histidine in the A1 type and proline in A2 β-casein. The presence of histidine in the A1 type increases the protein’s susceptibility to cleavage of the preceding seven amino acids, yielding β-casomorphin-7 (BCM-7), an exorphin with moderate agonistic activity on μ-receptors [1–3]. The knowledge that A1 β-casein may yield a functionally active exorphin has propelled research to understand whether A1 β-casein and BCM-7 affect gastrointestinal function or cause/exacerbate gastrointestinal inflammation in cell/animal models [4–9] and in people with milk or lactose intolerance [10, 11].

Thus far, however, very few studies have compared the effects of A1 and A2 β-casein on the gastrointestinal system in humans. In one recent study by Jianqin et al., 45 Chinese adults with self-reported intolerance to milk were randomised to receive milk containing either A2 β-casein or milk containing A1 and A2 β-casein (ratio 40:60; conventional milk) in a double-blind, 2×2 crossover manner with a washout period of 2 weeks before the study and between each phase [11, 12]. In that study, consumption of conventional milk was associated with increases in symptoms of post-dairy digestive discomfort and longer gastrointestinal transit times compared with milk containing only A2 β-casein. Consumption of conventional milk was also associated with increased small intestine inflammation and increased serum inflammatory biomarkers, as well as increased serum BCM-7 concentrations. In addition, when the subjects were divided according to their lactase activity based on urinary galactose tests, subjects with lactase deficiency reported greater gastrointestinal disturbances after consuming conventional milk compared with subjects with normal lactase activity, and that neither subgroup reported worsening of symptoms when they consumed milk containing A2 β-casein. Lactase deficiency was confirmed based on the subject’s reduced ability to metabolise lactose.

The results of this study prompted the hypothesis that the acute symptoms of milk intolerance (including self-reported lactose intolerance) in some people might be related to the presence of A1 β-casein in cow’s milk and that eliminating A1 β-casein could avoid these symptoms. Accordingly, the present study was designed to further evaluate the results of the study by Jianqin et al. [11, 12] by enrolling a much larger sample of subjects with self-symptoms of milk intolerance. Our primary objective was to compare the acute effects of milk containing either A2 β-casein or both A1 and A2 β-casein (conventional milk) on gastrointestinal symptoms in subjects with self-reported lactose intolerance and gastrointestinal discomfort. Our secondary objective was to examine whether these symptoms were correlated with lactase activity or the ability of these subjects to metabolise lactose.

Ho et al. [10] and Jianqin et al. [11] performed preliminary studies to compare the effects of conventional milk and milk containing only A2 β-casein on gastrointestinal symptoms in humans. Ho et al. [10] revealed that milk containing A1 β-casein was associated with significantly softer stool showing higher consistency scores, as determined using the Bristol Stool Scale, compared with milk containing A2 β-casein. In addition, consumption of A1 β-casein milk was associated with increased faecal calprotectin, a marker of intestinal inflammation [16]. Meanwhile, Jianqin et al. [11] revealed that consumption of conventional milk was associated with greater symptoms of post-dairy digestive discomfort in subjects with self-reported lactose intolerance. The worsening of gastrointestinal symptoms was apparent in lactose tolerant and lactose intolerant subjects. A subsequent analysis [17] of the study by Jianqin et al. revealed increased concentrations of inflammatory biomarkers and BCM-7 after consumption of milk containing both β-casein types compared with consumption of milk containing only A2 β-casein. However, the studies by Ho et al. [10] and Jianqin et al. [11] were relatively small, involving 40 and 45 subjects, and warranted confirmation in larger-scale studies. Nevertheless, the results highlighted a link between A1 β-casein, gastrointestinal inflammation, and symptoms of milk intolerance. Notably, subjects confirmed to be lactose malabsorbers tolerate milk containing only A2 β-casein, even though the lactose level was similar to that of conventional milk, suggesting that the type of β-casein may contribute to the symptoms of lactose intolerance in some people.

Accordingly, the objectives of the present were to compare the effects of consuming milk containing either A2 β-casein or conventional milk containing both A1 and A2 β-casein on acute self-recorded lactose intolerance and gastrointestinal discomfort occurring within several hours of consuming milk. In addition, we sought to examine the effects of both milk products on lactase activity to determine if changes in lactase activity are linked to the changes in self-reported symptoms of milk intolerance. We also examined whether age was correlated with a shift in lactase activity and the symptoms of milk intolerance.

This cross-over study of 600 Chinese subjects with self-reported milk intolerance revealed significant differences in gastrointestinal symptoms after the consumption of milk containing A2 β-casein or conventional milk. Of note, the gastrointestinal symptom scores were significantly lower at 1, 3 and 12 h after consumption of milk containing A2 β-casein relative to the consumption of conventional milk. These results suggest that elimination of A1 β-casein from the diet was associated with reduced severity of acute gastrointestinal symptoms after milk intake in this population.

It is important to note that the baseline symptoms were evaluated before consumption of either milk product by asking the subjects to report their symptoms at the last time they consumed milk. Accordingly, the subjects possibly recalled their worst experience. To avoid this potential source of bias, the analyses of gastrointestinal symptoms were adjusted for baseline scores to account for individual differences.

The exact mechanism by which acute exposure to A1 β-casein augments gastrointestinal symptoms relative to exposure to A2 β-casein is unclear, but we speculate that inflammation might be a contributing factor. This is supported by the studies by Ho et al. [10], Deth et al. [17], and Trivedi et al. [18] who noted increases in the concentrations of inflammatory biomarkers following exposure to A1 β-casein. However, these studies involved longer durations of exposure than our study, in which symptoms were assessed up to 12 h after exposure. To our knowledge, no studies have examined the acute effects of A1 β-casein exposure on gastrointestinal inflammation in humans.

Although no studies have examined the acute effects of A1 β-casein, some studies have investigated the acute effects of other dietary proteins on inflammatory biomarkers.

For example, Kristjánsson et al. [19] investigated mucosal inflammatory reactivity to cow’s milk protein and wheat gluten in 20 patients with coeliac disease and 15 healthy controls. The mucosal reactions to these proteins were assessed 15 h after exposure. Of note, the gluten challenge induced neutrophil activation and nitric oxide synthesis. Ten patients showed strong inflammatory reactions to cow’s milk protein. Six patients sensitive to cow’s milk were also challenged with casein and α-lactalbumin. In this experiment, casein induced an inflammatory response similar to that elicited by cow’s milk. These findings suggest that casein elicits an inflammatory response similar to that elicited by gluten in patients with coeliac disease. These results are consistent with the study by Trivedi et al. [18] who reported that A1 β-casein-derived BCM-7 and gluten-derived exorphin share a mechanistic pathway for inducing oxidative stress in cultured human gut epithelial cells and neuronal cells.

Holmer-Jensen et al. [20] conducted a randomized crossover study in which 11 obese non-diabetic subjects consumed a fat-rich mixed meal containing cod protein, whey isolate, gluten, or casein. They observed some differences in the acute effects of dietary protein on postprandial inflammatory biomarkers. Intriguingly, all four proteins were associated with reductions in monocyte chemoattractant protein-1 and increases in CCL5/RANTES. The whey protein meal was associated with the smallest reduction in monocyte chemoattractant protein-1 and the largest increase in CCL5/RANTES compared with the other meals.

Pal and Ellis [21] compared the effects (within 6 h) of whey protein, caseinate, and glucose on blood pressure, vascular function, and inflammatory markers in 20 overweight and obese postmenopausal women. Although systolic blood pressure, diastolic blood pressure, and augmentation index decreased initially after each meal, there were no significant differences in these variables between the glucose, casein, or whey groups. Moreover, they found no differences in plasma inflammatory markers.

Finally, Nestel et al. [22] found no changes in systemic inflammatory and atherogenic biomarkers after ingestion of a variety of dairy products (low-fat milk, or 45 g fat from butter, cream, yoghurt, or cheese) in 12 overweight subjects after a single meal. Moreover, in a 4-week study of 12 subjects who consumed 50 g dairy fat daily as either butter, cream and ice cream (non-fermented) or cheese plus yoghurt (fermented) dairy foods, there were no apparent differences in fasting biomarker concentrations between the non-fermented and fermented dairy products.

Unfortunately, none of these studies assessed gastrointestinal symptoms and changes in plasma inflammatory markers might not be correlated with local inflammation.

Nevertheless, the results of these studies suggest that dietary proteins might have differential effects on gastrointestinal inflammation, and further studies might be necessary to examine whether changes in localised gastrointestinal inflammation are correlated with gastrointestinal symptoms.

It is also important to consider that lactose might contribute to the gastrointestinal symptoms in this cohort of subjects with self-reported lactose intolerant. Indeed, an increase in gastrointestinal symptoms was observed when the subjects consumed conventional milk. However, the symptoms were reduced when the subjects consumed milk containing only A2 β-casein, indicating that A1 β-casein-induced inflammation may be linked to the symptoms of lactose intolerance.

To examine the impact of lactose malabsorption on gastrointestinal symptoms, we divided the subjects as lactose absorbers and lactose malabsorbers, based on the results of the urinary galactose test. Of note, the gastrointestinal symptoms after consumption of milk containing A2 β-casein were comparable between the lactose absorbers and lactose malabsorbers.

Based on these findings, we propose the hypothesis that the gastrointestinal symptoms in some subjects with self-reported lactose intolerance might be related to A1 β-casein rather than lactose itself. This seems feasible considering that the lactose concentrations were comparable in both milk products.

We also explored the possibility that age had an impact on gastrointestinal symptoms or the correlation between lactose malabsorption and gastrointestinal symptoms. As indicated in Additional file 1: Tables S2, S5 and S6, age was not significantly associated with gastrointestinal symptoms. However, because the upper age range was limited to 50 years, it is possible that older subjects might experience more severe gastrointestinal symptoms after dairy intake.

The results of this study should be interpreted with care, considering the limitations of this study, especially in terms of the mechanistic link between the observed compromise in lactose digestion and the type of β-casein. In addition, we used an indirect method to assess lactase activity. Last, the consumption of other non-dairy foods and drinks by subjects is a potential confounder; however, rather than deny intake of food to participants, we ensured that any foods and drinks consumed were dairy-free and that there was consistency in food intake type for both interventions. Further studies are warranted to examine the putative role of A1 β-casein in gastrointestinal inflammation, the effects of inflammation on the expression and/or activity of lactase enzyme, and the proportion of people with lactose intolerance who would benefit from excluding A1 β-casein from their diet. Additionally the effects of long-term exposure to milk in terms of the changes in gastrointestinal health need to be examined in future trials, and whether chronic exposure conditions desensitise the gastrointestinal tract to A1 β-casein consumption.

In conclusion, this study showed that consumption of milk containing A2 β-casein attenuated the acute gastrointestinal symptoms following milk intake relative to conventional milk containing A1 and A2 β-casein in Chinese subjects with self-reported lactose intolerance. Gastrointestinal symptoms after consuming milk containing A2 β-casein were consistently reduced in both lactose absorbers and lactose malabsorbers. These findings suggest that, in some individuals with self-reported lactose intolerance, the adverse gastrointestinal symptoms following milk intake might be related to the presence of A1 β-casein in milk rather than lactose itself.