Food allergies occur in 5% to 10% of infants and young children (1,2). Cow milk protein allergy (CMPA) is the most common sensitivity of young infants, with a 2- 6% incidence (3,4). This atopic disease is associated with a broad spectrum of immunoglobulin (Ig) E-mediated reactions that are mostly expressed as immediate symptoms, such as urticaria, rhinoconjunctivitis, asthma, vomiting, diarrhea and sometimes systemic anaphylactic shock and death (5).
CMPA may not be an isolated phenomenon, but may herald the beginning of an “atopic career.” Infants and children with CMPA may also develop other atopic diseases such as other food allergies, rhinitis, asthma or early atopic dermatitis, a condition affecting up to 15-20% of normal children (6–10).
Human milk contains numerous factors protective against atopic disorders (11,12). However, breast-feeding is not always possible and in most cases children receive breast milk for less than 6 months. At the end of the breast-feeding period, children are usually given a cow milk based formula and at this time some become sensitized to cow milk (CM) and develop immediate symptoms after ingestion of cow milk. For these children a milk substitute must be provided. The most common alternatives to cow milk are soy milk and hydrolyzed formulas. There is, however, evidence that 10% to 40% of children allergic to CMP do not tolerate soy derivatives (13–15). Hydrolyzed formulas are usually tolerated by allergic children but some studies report that these formulas fail to induce a desired oral tolerance in rodents (16–18). In addition, they lack some of the bioactive peptides present in untreated cow milk that play important roles in immunomodulation (19).
The identification of a more suitable protein source for cow milk allergic children represents an important goal for pediatricians and nutritionists. Goat milk (GM) has been proposed as an alternative to cow milk because some allergic children can tolerate goat milk derivatives (20,21). However, several studies demonstrate that children with CMPA are also sensitized to goat milk protein, suggesting cross-reactivity between cow and goat milk proteins (22–24). There are rare reports of patients allergic to goat but not to cow milk proteins (25). These reports suggest that goat milk may not be a suitable protein source for children with CMPA.
There is evidence, nevertheless, to believe that GM is less allergenic than CM because the content in α-casein, one of the most important antigens in CM, is lower in GM (26). It has been reported that α-casein acts as a carrier for other allergens in cow milk such as β-lacto-globulin, which is tightly linked to the casein micelles and therefore more difficult to digest (27). The lower content of α-casein in GM might allow a better digestion of β-lactoglobulin and other allergens. Therefore, GM could be less immunogenic when introduced into the diet of non-sensitized children. In this study we compared the allergenicity of CM and GM in a mouse model of orally induced atopy.
MATERIALS AND METHODS
Reagents and Milk Proteins
Cholera toxin (CT), concanavalin A (Con A) and lipopoly-saccharide (LPS) were all purchased from Sigma Chemical Co. (St Louis, MO). Antibodies for ELISAs were purchased from Bethyl (Montgomery, TX) in the case of immunoglobulins or from Biosource (Camarillo, CA) for cytokines. All the other products were of the best grade available and purchased from Sigma. Deionized water, further purified with a Millipore Milli-Q (Bedford, MA) system, was used.
Homogenized CM was obtained from Puleva Food S.A. (Granada, Spain) (protein composition in g/liter: total protein: 36, total casein 29.5, αs1 casein: 11.9, αs2 casein: 3.1, β casein:9.8) whereas GM was obtained fresh from murciano-granadina dairy goats in a local farm (protein composition in g/L: total protein: 33, total casein 28, αs1 casein: 1.4, αs2 casein: 7, β casein: 18.9) and submitted to the same UHT treatment as CM. Whole milk thus obtained was centrifuged at 3500 rpm for 10 minutes at 4°C and the upper layer of fat was discarded to obtain a skimmed milk. The content of protein in CM and GM was measured by the Kjeldhal method as previously described (28). Both milks were stored at -20°C in 1-mL aliquots and defrosted immediately before use in experiments.
Just weaned, 3 week old female Balb/C mice were purchased from the Granada University breeding colony and housed in a light controlled environment (12 hour cycle of light/dark) at 22°C. Animals (13 per group) were maintained on a regular mouse chow diet under pathogen-free conditions. Guidelines for the care and use of animals were followed as described (29). The entire experiment was conducted three times with similar results.
Sensitization and Challenge by Oral Administration of Antigen
Mice were sensitized intragastrically with CM or GM plus CT as an adjuvant and boosted five times at weekly intervals. Intragastric feeding was performed by means of a PVC tube feeding needle purchased from Vygon (Ecoue, France). Mice received 1 mg of CMP (CM group) or GMP (GM group) per gram of body weight together with 0.3 μg/g of CT. The effect of the adjuvant per se was evaluated using two control groups, one receiving CT in phosphate buffer saline (PBS) (CT group) and another group receiving only PBS (C group). Six weeks after the first feeding mice were fasted overnight and challenged intragastrically. The C, CT and CM groups received two doses of CMP (30 mg/mouse) given 30 minutes apart. Similarly the GM group received two doses of GMP (30 mg/mouse) given 30 minutes apart. Previous experiments performed by our group had shown that control groups (C, CT) could be challenged either with CM or GM with identical results (data not shown).
Thirty minutes after the last challenge, mice were killed by intraperitoneal administration of sodium pentothal (50 mg/kg), and blood was collected by cardiac puncture in ethylenediaminetetraacetic acid containing tubes. Spleen and intestine were removed from each mouse and weighed.
Evaluation of Symptoms
Diarrhea was used as the gastrointestinal symptom of interest. Diarrhea usually occurred within 7 days of the second dose. Animals suffering from diarrhea in at least two consecutive weeks were considered positive cases.
Immediately after challenge, hypersensitivity responses were evaluated by using a scoring system modified slightly from previous reports as follows: 0 = no symptoms; 1 = scratching and rubbing around the nose and head; 2 = puffiness around the eyes and mouth and pilar erecti; 3 = decreased activity with increased respiratory rate; 4 = wheezing and labored respiration; 5 = cyanosis around the mouth and the tail; 6 = death (30).
Evaluation of symptoms was performed by two independent investigators who were not directly involved in the study and who were unaware of the animals’ experimental group.
Measurement of CM or GM-Specific IgG1 in Plasma
Blood was centrifuged at 3500 rpm for 10 minutes at 4°C and plasma aliquots were collected and frozen at -80°C. Levels of CM or GM-specific IgG1 were measured as previously described (30). Briefly, 96-well plates were coated with 20 μg/mL of protein from cow or goat milk in coating buffer 0.5 M). After overnight incubation at 4°C, plates (Na2CO3 were washed three times with wash solution (50 mM Tris, 0.14M NaCl, 0.05% Tween 20) and blocked with blocking solution (50 mM Tris, 0.14 NaCl, 1% BSA). Then, plasma samples (1:10 dilutions) were added to the plates and incubated for 1 hour at room temperature. Plates were then washed three times and 100 μL of goat anti-mouse IgG1 antibody conjugated with peroxidase (Bethyl Laboratories) was added for an additional 1 h at room temperature. Staining was performed with TMB (Sigma Chemical) for 30 minutes at room temperature and in the dark, stopped with 0.1N H2SO4 and plates were read at 450 nm. Results were expressed as optical density (OD) ± SD. All analyses were performed in duplicate.
Determination of Plasma Histamine Levels
Plasma histamine levels were determined by an enzyme immunoassay kit (IBL Laboratories, Hamburg, Germany) following manufacturer recommendations.
Spleen Cells Culture
Spleens were removed from all mice 30 minutes after the last challenge and homogenized in complete culture medium (DMEM plus 10% fetal bovine serum [FBS], 1% penicillin/streptomycin). After centrifugation at 1500 rpm for 5 minutes erythrocytes were lysed with a lysing solution (NH4Cl, 1.7 M; KHCO3, 0.12 M; EDTA, 0.009 M) for 15 minutes at 4°C. Resting cells were counted by using a hemacytometer and cultured to perform proliferation and stimulation assays in current culture medium (DMEM + 10%FBS). Cells were incubated at 37° C in a humidified 5% CO2 atm.
Spleen derived lymphocytes were cultured in 24-well plates (2 × 106 cells/well) in 1 mL of medium. Lymphocytes from every mouse were divided into three wells, one incubated in the absence of proteins (basal), one in the presence of CMP (100 μg/mL) and one in the presence of GMP (100 μg/mL). [3H]-tymidine (1 μCi/mL) (Amersham Biosciences, Arlington Heights, IL) was added to each well. After 48 h plates were centrifuged at 1500 rpm for 5 minutes and supernatants were discarded. To fix cells 500 μl of ice-cold 70% methanol was added to each well. After three washes in ice-cold 10% trichloroacetic acid, cells were solubilized in 1% SDS and 0.3 M NaOH at room temperature. Radioactivity was counted by liquid scintillation using a 2100 Tri-Carb Packard (Meriden, CT) scintillation counter. Each point was performed in duplicate and the results were expressed as the mean ± SD.
Analysis of Cytokine Production
Spleen derived lymphocytes were cultured in six-well plates (10 × 106 cells/well) in 5 mL of medium and incubated during 24 h in the presence or absence of ConA (5 μg/mL), LPS (50 μg/mL), CMP or GMP (100 μg/mL). The supernatants were collected after 48 hours of incubation. Levels of IgA, IgG1, interleukin (IL)-2, IL-4, tumor necrosis factor-μ and interferon (IFN)- μwere determined by ELISA according to the manufacturer instructions (Bethyl and Biosource).
Analysis of Gene Expression by Reverse Transcription-Polymerase Chain Reaction
Total mRNA was isolated from cultured spleen cells by using Trizol Reagent (Invitrogen, Carlsbad, CA) as described by the manufacturer. The reversed transcription was performed by using the Reverse Transcriptase AMV (Roche, Indianapolis, IN) as described elsewhere (31). First-strand cDNAs were either stored at -20°C or used for the polymerase chain reaction (PCR) step. PCR reactions (25 μ L total volume) were performed as previously reported (31). Once the PCR reactions were completed, 20 μ L of each reaction mixture was used for electrophoresis on 2% agarose gels, and DNA bands were visualized with ethidium bromide staining and ultraviolet irradiation. Murine oligonucleotides primers used in the PCR were purchased from MWG-Biotech AG (Ebersberg, Germany) and the nucleotide sequences were as follows: IFN-γ, forward 5′-TGGAGGAACTGGCAAAAGGATGGT-3′, reverse 5′-TTGGCACAATCTCTTCAC-3′; IL-4, forward 5′-ACGA-GGTCACAGGAGAAGGGA-3′, reverse 5′-GGAGCA-GCTTATCGATGAATC-3′ β-actin, forward 5′-TGGAATC-CTGTGGCATCC-3′, reverse 5′-AACGCAGCTCAGTAA-CAGTCC-3′.
Statistical significance (P < 0.05) was calculated by Student’s t-test in case of parametric parameters such as the results of ELISAs and proliferation assays. All tests were performed with one tail following the two-sample equal variance model (homoscedastic). For nonparametric parameters (score, intestine weight/body weight ratio and number of mice with diarrhea) Mann-Whitney U-test or Fisher’s exact test were used to determine statistical significance (P < 0.05). Finally, the Spearman test was used for the statistical analysis of the correlation between antigen-specific IgG1, serum histamine concentration and hypersensitivity score.
The percent of animals with diarrhea was higher in CM-sensitized than GM-sensitized mice (P = 0.02 versus CM) (Table 1). Diarrhea in neither group was so severe as to cause a significant decrease in body weight or in food intake (data not shown). We did not expect more severe symptoms because this is a model of milk atopy rather than a model of milk allergy.
After 6 weeks of sensitization, mice were challenged intragastrically with two doses of 30 mg/mouse of CM (C, CT and CM groups) or GM (GM group) without cholera toxin. Hypersensitivity symptoms became evident within 15 to 30 minutes. The severity of the reaction was scored as indicated above. As with diarrhea, the most severe allergic reactions were also observed in CM-sensitized mice (Fig. 1). Mice sensitized with GM showed weaker reactions and control mice showed no symptoms of hypersensitivity after CM challenge. Mice of the CT-group had a mild reaction that was probably attributable to weak allergic responses to allergens in food as a consequence of the immunostimulation caused by cholera toxin. The average severity score of the reaction in the CM group was 2.923 ± 1.115 (mean ± SD), which was significantly higher (P < 0.01) than that observed in the CT control group (0.615 ± 0.506). In con- trast, the score of the GM group was 1.153 ± 1.068, which was significantly lower (P < 0.01) than that of the CM group but did not statistically differ from that of the CT control group (P = 0.206) (Fig. 1). Taken together, these results indicate that we achieved a sensitization that was clearly less severe in the GM group.
Thirty minutes after the second dose of challenge, mice were killed under anesthesia and the intestine of each mouse was removed and weighed. The ratio of intestine weight to body weight was significantly higher in the CM group (0.113 ± 0.008) compared with the GM group (0.100 ± 0.009) (P = 0.02) and the CT group (0.98 ± 0.004) (P = 0.01) although no edema nor ulceration was grossly observed. This might indicate a weak intestinal inflammation caused by CM sensitization consistent with the gastrointestinal symptoms of the CM group.
Because lymphocytes play a key role in the development of CMPA, the spleen of each mouse was removed and weighed. Spleen-derived lymphocytes were counted and cultured to perform proliferation assays and to study their response to different stimuli. Lymphocytes from CM-sensitized mice significantly increased their proliferation ratio when CMP was added to the culture medium (Fig. 2). In contrast, GMP failed to induce a significant increase in the proliferation ratio of lymphocytes from the GM group. However, this increase in the proliferation ratio in the CM group was not reflected in spleen weight and number of cells, which were similar in all groups (data not shown), suggesting that the number of lymphocytes sensitized to CMP represents a low percentage of total spleen cells.
The production of cytokines by spleen cells derived from CM-sensitized and GM-sensitized mice was analyzed to investigate what type of lymphocyte response was responsible for the development of milk atopy. Spleen-derived lymphocytes were cultured in the presence or absence of a nonspecific stimulus (ConA). After 48 hours in culture, IL-4 levels were significantly increased in lymphocytes from CM-sensitized mice, compared with those from control mice and GM-sensitized mice (P < 0.05) (Fig. 3A). Furthermore, IFN-β production by spleen cells from CM-sensitized mice was significantly lower than those from GM-sensitized mice (P < 0.05) (Fig. 3C).
Consistent with these results, the analysis of cytokine expression by reverse transcription-PCR showed an increased expression of IL-4 (Fig. 3B) and a decreased expression of IFN-β in spleen cells from CM-sensitized mice (Fig. 3D). In all cases GM-sensitized mice showed a response similar to that of the control groups.
To study systemic effects of sensitization, plasma was collected from each mouse immediately after killing. Plasma levels of antigen-specific IgG1 were measured by ELISA. As shown in Figure 4A, CM induced a higher synthesis of CMP-specific IgG1 (P < 0.05) compared with mice from the CT and C control groups. In contrast, GM did not induce such an IgG1 response, as GMP-specific IgG1 levels were similar to those of the CT-treated and control mice. Antigen-specific IgE levels in the same samples were too low to be detected by the ELISA method used (data not shown). Moreover, no differences were observed in total plasma amounts of IgG, IgG1 or IgE in all groups (data not shown).
Plasma histamine levels were significantly increased in CM-sensitized mice when compared with CT treated mice (Fig. 4B). However, the plasma histamine concentration in GM-sensitized mice was similar to that of the control groups. We observed that mice with a higher score had also the highest levels of histamine liberation (P < 0.01, r = 0.520, n = 49) (Fig. 5), thus suggesting that the higher histamine liberation was responsible for the symptoms of allergy.
To our knowledge this is the first animal study comparing cow milk and goat milk as first protein sources after breast-feeding. Although no ideal model exists, we used a previously described animal model (30) that exhibits the characteristics of type 1 hypersensitivity reactions to compare the allergenicity of cow milk and goat milk. The purpose of using this model was to mimic human food allergy by provoking food hypersensitivity by oral ingestion. Other animal models have not been reported to provoke systemic reactions (32,33) or have been induced by parenteral challenge (34), so they did not adequately mimic human food allergy.
To bypass the tendency of mice to develop oral tolerance, we used CT, which has been reported to stimulate a Th2 response and the production of IgG1 antibodies (35,36). Furthermore, Balb/C mice were used because this strain of mice is prone to develop Th2 responses easier than other strains (37).
In the animal model used in this study IgG1 seems to be responsible for mast degranulation and histamine release. This is in agreement with previous studies demonstrating that IgG1 can mediate type I hypersensitivity in mice (38,39). It is well known that mouse IgG1 binds to Fc-receptors in mast cells and promotes degranulation when reacting against a specific antigen, thus increasing histamine levels. This increase in histamine liberation is probably responsible for the symptoms observed in this study. Whereas CM provoked a strong IgG1 response after intragastric challenge, GM failed to induce such a response. Thus histamine release was higher in CM-sensitized mice and was probably responsible for hypersensitivity symptoms.
T cells and cytokines play an important role in regulating allergic diseases such as asthma (40,41). However, the extent of T cells involvement in food allergy is unclear. In our study, CM stimulation of spleen cells from mice allergic to CM induced an increase in the proliferation ratio of spleen cells. It also induced a significant increase in the synthesis of IL-4 but a decrease in the production of IFN-γ, both crucial cytokines for Th2 and Th1 responses, respectively. These findings suggest that development of CMPA might be mediated by a Th2 response, as previously reported (30). In CM-sensitized mice the increase in IL-4 production might stimulate IgG1 synthesis by specific B cells, thus explaining the higher levels of CMP-specific IgG1 levels in CM-sensitized mice. In contrast, GM did not induce a similar response, thus suggesting that GM-sensitized mice developed a less severe hypersensitivity reaction.
In conclusion, we compared CM and GM as first sources of protein after weaning in a murine model of food allergy. Our results suggest that GM is less immunogenic than CM because GM induced a weaker Th2 response. Further studies are needed to determine whether GM could be a suitable substitute for CM in infant formulas.
The authors are grateful to Mr. Arjan Geerlings for English revision and Mr. Juan Antonio Maldonado for his kind help in the statistical analysis.
1. Bock SA. Prospective appraisal of complaints of adverse reactions to food in children during the first three years of life. Pediatrics
2. Sampson HA. Immediate reactions to food in infants and children. In: Metcalfe DD, Sampson HA, Simon RA, eds. Food allergy: adverse reactions to food and food additives
. Boston: Blackwell Scientific Publications, 1997:169.
3. Hill DJ. Cow’s milk allergy in infancy and early childhood. Clin Exp Allergy
4. Hosking CS, Heine RG, Hill DJ. The Melbourne milk allergy study: two decades of clinical research. Allergy Clin Immunol Int
5. Sampson HA. Food allergy. JAMA
6. Oehling A, Resano A, Sanz ML, et al. Importance of food allergy in atopic dermatitis. Allergy
7. Wüthrich B. Epidemiology and natural history of atopic dermatitis. Allergy Clin Immunol Int
8. Schäfer T, Vieluf D, Behrendt H, et al. Atopic eczema and other manifestations of atopy: results of a study in East and West Germany. Allergy
9. Businco L, Bartolucci M. Atopic dermatitis and food allergy in Europe-prevalence and risk factors. Allergy
10. Broberg A, Svensson A, Borres MP, et al. Atopic dermatitis in 5–6 year old Swedish children: cumulative incidence, point prevalence and severity scoring. Allergy
11. Lovegrove JA, Hampton S, Morgan JB. The immunological and long-term pregnancy and lactation: a pilot study. Br J Nutr
12. Lonnerdal B. Nutritional and physiologic significance of human milk proteins. Am J Clin Nutr
13. Hill DJ, Ford RP, Shelton MJ, et al. A study of 100 infants and young children with cow’s milk allergy. Clin Rev Allergy
14. Chandra RK, Puri S, Hamed A. Influence of maternal diet during lactation and use of formula feeds on development of atopic eczema in high risk infants. Br Med J
15. Perkkio M, Savialhti E, Kuitunen P. Morphometric and immunohistochemical study of jejunal biopsies from children with intestinal soy allergy. Eur J Pediatr
16. Fritsché R, Pahud JJ, Pecquet S, et al. Induction of oral tolerance to cow’s milk proteins in rats fed with protein hydrolysate. Nutr Res
17. Fritsché R, Pahud JJ, Pecquet S, et al. Induction of systemic immunological tolerance to beta-lactoglobulin by oral administration of a whey protein hydrolysate. J Allergy Clin Immunol
18. Pecquet S, Bovetto L, Maynard F, et al. Peptides obtained by tryptic hydrolysis of bovine beta-lactoglobulin induce specific oral tolerance in mice. J Allergy Clin Immunol
19. Harsharnjit S, Doull F, Rutherfurd KJ, et al. Immunoregulatory peptides in bovine milk. Br J Nutr
2000;84 Suppl 1:S111–7.
20. Ellis MH, Short J, Heiner DC. Anaphylaxis after ingestion of a recently introduced hydrolysed whey formula. J Pediatr
1991;118: 74 –7.
21. Webber SA, Graham Brown RAC, Hutchinson PE, et al. Dietary manipulation in childhood atopic dermatitis. Br J Dermatol
22. Bernard H, Wal JM, Creminon C, et al. Sensitivities of cow’s milk allergic patients to casein fractions of milks from different species (abstract). Allergy
23. Spuergin P, Walter M, Chiltz E, et al. Allergenicity of alpha-caseins from cow, sheep and goat. Allergy
24. Carroccio A, Cavataio F, Iacono G. Cross-reactivity between milk proteins of different animals. Clin Exp Allergy
25. Umpierrez A, Quirce S, Maranon F, et al. Allergy to goat and sheep cheese with good tolerance to cow cheese. Clin Exp Allergy
26. Bellioni-Businco B, Paganelli R, Lucenti P, et al. Allergenicity of goat’s milk in children with cow’s milk allergy. J Allergy Clin Immunol
27. Bevilacqua C, Martin P, Chándal C, et al. Goat’s milk of defective α s1-casein genotype decreases intestinal and systemic sensitization
to β-lactoglobulin in guinea pigs. J Dairy Res
28. Association for Official Analytical Chemists (AOAC). Official methods of analysis
, 13th ed. Washington, DC: Association for Official Analytical Chemists, 1980:777.
29. Institute of Laboratory Animal Resource Commission of Life Sciences. National Research Council. Guide for the care and use of laboratory animals
. Washington, DC: National Academy Press, 1996.
30. Xiu-min Li. Brian H Schofield, Ching-Kang Huang, et al. A murine model of IgE-mediated cow’s milk hypersensitivity. J Allergy Clin Immunol
31. Soler C, Garcia-Manteiga J, Valdes R. Macrophages require different nucleoside transport systems for proliferation and activation. FASEB J
32. Boza JJ, Jimenez J, Martinez O, et al. Nutritional value and antigenicity of two milk protein hydrolysates in rats and guinea pigs. J Nutr
33. Boza JJ, Martinez-Augustin O, Gil A. Nutritional and antigenic characterization of an enzymatic whey protein hydrolysate. J Agric Food Chem
34. Poulsen OM, Hau H, Kollerup J. Effect of homogenization and pasteurization on the allergenicity of bovine milk analysed by a murine anaphylactic shock model. Clin Allergy
35. Snider DP, Marshall IS, Perdue MH, et al. Production of IgE antibody and allergic sensitization
of intestinal and peripheral tissues after oral immunization with protein Ag and cholera toxin. J Immunol
36. Marinaro M, Staas HF, Hitol T, et al. Mucosal adjuvant effects of cholera toxin in mice results from induction of T helper 2 (Th2) cells and IL-4. J Immunol
37. Sun S, Winship T, Kuchan MJ. Dietary ribonucleotides modulate type 1 and type 2 responses against OVA in young BALB/c mice. J Nutr
38. Oettgen HC, Martin TR, Wynshaw-Boris A, et al. Active anaphylaxis in IgE-deficient mice. Nature
39. Li XM, Kleiner G, Huang CK, et al. Murine model of atopic dermatitis associated with food hypersensitivity. J Allergy Clin Immunol
40. Robinson D, Hamid Q, Bentley A, et al. Activation of CD4+
T cells increased Th2-tyope cytokine mRNA expression and eosinophil recruitment in bronchoalveolar lavage after allergen inhalation challenge in patients with atopic asthma. J Allergy Clin Immunol
41. Lukacs NW, Streiter RM, Chonsue SW, et al. Interleukin-4-dependent pulmonary eosinophil infiltration in a murine model of asthma. Am J Respir Cell Mol Biol