- Email: [email protected]
hypersensitivity: Antigenicity or timing?
ARTICLE IN PRESS
Immunobiology 214 (2009) 269–278 www.elsevier.de/imbio
Food allergy/hypersensitivity: Antigenicity or timing? Patrı´cia Olaya Paschoala,b,, Sylvia M.N. Camposb, Monique M.B. Pedruzzib, Vale´ria Garridob, Moˆnica Bissoa, Danielle M.F. Antunesb, Alberto F. Nobregac, Gerlinde Teixeirab a
Programa de Po´s-Graduac¸a˜o em Patologia, Faculdade de Medicina, Universidade Federal Fluminense, Brazil Laborato´rio do Grupo de Imunologia Gastrintestinal, Departamento de Imunobiologia, Instituto de Biologia, Universidade Federal Fluminense, Brazil c Universidade Federal do Rio de Janeiro b
Received 30 October 2007; received in revised form 25 August 2008; accepted 15 September 2008
Abstract Mechanisms involved in the induction of oral tolerance (OT) or systemic immunization through the oral rout are still poorly understood. In our previous studies, we have shown that when normal mice eat peanuts they become tolerant, with no gut alterations. Conversely, if immunized with peanut proteins prior to a challenge diet (CD) containing peanuts they develop chronic inflammation of the gut. Our aim is to evaluate the consequences of the introduction of a novel protein in the diet of animals presenting antigen-specific gut inflammation. Adult, female C57BL/6J mice were divided in control (C) and experimental (E) groups. C1–C3 received peanut protein immunization, animals of the control groups C4 were sham immunized, and control group C5 received ovalbumin (OVA) immunization. The experimental group was immunized with peanut protein extract. Before initial exposure to a 30-day peanut containing CD, the experimental group was divided into 5 groups (E1–E5). OVA feeding began 7 days prior CD (E1) on day 0 (E2), 7 (E3), 14 (E4) and 21 (E5) during CD. Our results show that oral exposure to a novel protein (OVA) in the absence of gut inflammation (E1) leads to low levels of systemic antibody (Ab) titers, comparable to tolerant animals. Conversely, as off initial induction of inflammation, groups submitted to OVA (OT) protocol develop increasingly higher systemic Ab titers similar to animals of the immune control group. In conclusion, our protocol indicates that timing is more important than the antigenicity when a novel protein is offered, in the diet. r 2008 Elsevier GmbH. All rights reserved. Keywords: Food allergy; Ovalbumin; Oral tolerance; Peanuts
Introduction For many years the term allergy was used as a synonym for IgE-mediated hypersensitivity, thus Corresponding author at: Instituto de Biologia, Alameda Barros Terra s/n, Centro, Nitero´i, 24020-150, Rio de Janeiro, Brazil. Tel.: 55 21 26292313. E-mail address: [email protected] (P.O. Paschoal).
0171-2985/$ - see front matter r 2008 Elsevier GmbH. All rights reserved. doi:10.1016/j.imbio.2008.09.007
associated to severe potentially life-threatening anaphylactic reactions. However, clinical observations have shown that not all allergies have this as the underlining mechanism. A force task of the European Academy of Allergy and Clinical Immunology published in 2001 a revised nomenclature for allergies (Johansson et al., 2001). This force task proposed that allergies are hypersensitivity reactions initiated by immunologic mechanisms and that adverse reactions to food should
ARTICLE IN PRESS 270
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
be called food hypersensitivity. When immunologic mechanisms have been demonstrated, the appropriate term is food allergy, and, if the role of IgE is highlighted, the term is IgE-mediated food allergy. All other reactions, previously referred to as ‘‘food intolerance’’, should be referred to as nonallergic food hypersensitivity as previously suggested by Bruijnzeel-Koomen et al. (1995) and Ortolani et al. (1999). In 1946, Merill Chase showed that oral administration of a contact-sensitizing agent (2,4-dinitrochlorobenzene) does not lead to sensitization, but rather prevents the animal from eliciting an immune response upon subsequent subcutaneous injections or cutaneous challenge. The specific induction of these regulated responses by administration of antigen through the gastrointestinal tract is known as oral tolerance (OT) (Mayer and Shao, 2004). Thus, food allergy represents an abrogation of normal OT to that specific foodstuff (Scott and Sampson, 2006). For efficient strategies aimed at an active cure of food allergies, a better understanding of the pathogenic mechanisms is strongly needed (Sampsom et al., 2005). In contrast to current thought that food allergy is primarily a Th2-driven process, there is accumulating data from the literature indicating that food allergies may also be a result of a mixed Th1Th2 immune response. The mucosal immune system utilizes a very complex and tightly regulated system of suppression and controlled inflammation to distinguish between innocuous and potentially harmful antigens of the intestinal lumen. Although the exact mechanisms governing this regulation remains to be elucidated, evidence exists for the involvement of a variety of cell types, including CD4+, CD8+ T cell, regulatory T cells, B cells, and dendritic cells (Mayer, 2005). When both the ability to remain tolerant to the abundance of dietary antigens and prevention of luminal bacterial flora to harm the host fail, disease may install (Kraus et al., 2004a). OT induction is a key feature of intestinal immunity, generating systemic low responsiveness to ingested antigens (Worbs et al., 2006). Mechanisms governing tolerance induction have been well characterized in selected mouse strains. Similar studies in man are lacking, although there is evidence that tolerance can be induced. The ability to induce OT in humans was reported by Husby et al., although the type of tolerance generated to a neo-antigen, keyhole limpet hemocyanin (KLH), was slightly different from that seen in murine models. Tolerance for delayed-type hypersensitivity (DTH) responses but not antibody (Ab) responses was seen (Husby et al., 1994; Kraus et al., 2004b). In disease states, tolerance is altered and this may account for the presence of mucosal inflammation. In food hypersensitivity there is evidence that allergens may be handled differently and this may play a role
in disease expression (Mayer and Shao, 2004). It has been clearly demonstrated that an upregulated Th1 immune response and subsequent overproduction of Th1 cytokines may lead to autoimmune diseases such as, inflammatory bowel disease, autoimmune arthritis, experimental autoimmune encephalomyelitis, and typeI diabetes (Kuchroo et al., 1995; Neurath et al., 2002). Thus, there is an intense interest in understanding the regulation of the immune response, with the goal of developing therapeutic strategies. In cases of multiple food allergies, restricted diets may result in unbalanced nutrition and pose a considerable hardship to the family of a food-allergic child (Nowak-Wegrzyn, 2003). In previous studies of our group, we have shown that crude peanut protein extract is immunogenic both in mice and rats. All animals responded with high levels of systemic antibodies (po0.001) as compared to saline controls. No differences were observed when comparing animals within strains nor when comparing young adults (2–4 months) with older animals (12–16 months). Submitting animals to a peanut containing challenge diet (CD) induces important alterations of the gut mucosal architecture, typical of chronic inflammation in immunized but not in normal animals (Teixeira, 2003; Teixeira et al., 2008a, in press). In this study, we show the consequences of the administration of a non related protein in the presence of a chronic antigen-specific gut inflammation.
Methods Animals C57BL/6J female mice, 2–3 months old, were obtained by the local Animal Resources Center (Universidade Federal Fluminense). These animals were randomly divided into 2 groups: control and experimental. These were subsequently subdivided into three control groups (n ¼ 7) and 5 experimental groups (n ¼ 14). All animals were individually marked, to be able to perform paired statistical analysis. This research was approved by the animal ethics committee of the School of Medicine of the Fluminense Federal University.
Immunization protocols The immunization protocol utilized in this study was: 100 mg specific protein (peanut protein extract or ovalbumin (OVA)) plus 1 mg of adjuvant [Al(OH)3], in a final volume of 200 ml by subcutaneous route (sc) for the primary immunization and 100 mg of specific protein without adjuvant, performed 21 days later, for the booster immunization.
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
For normal controls the animals were sham immunized with 1 mg of adjuvant [Al(OH)3], in a final volume of 200 ml of saline by sc for the primary immunization and saline without adjuvant, performed 21 days later, for the booster immunization. All animals of the experimental groups (E1–E5) and control groups C1–C3 received peanut protein immunization, animals of the control groups C4 were sham immunized, and control group C5 received OVA immunization.
Tolerization to OVA or peanuts To induce OT, animals received to drink a sweetened 20% egg white solution (diluted in saline v/v with 5% of saccharine) or to eat a diet containing exclusively peanut seeds for 7 days ad libitum prior to the immunization protocol described above with the respective proteins. (Teixeira, 2003; Teixeira et al., 2008a, in press).
271
Animals of control group immunized to peanuts (C1) continued to receive commercial chow, (C2) which had been submitted to peanuts OT protocol and (C3) immunized to peanuts received peanuts for a 30-day period.
Timing of OVA feeding in mice with antigen-specific inflammation of the gut To determine the consequences of gut inflammation on the induction of OT, the 5 experimental animals groups received OVA to drink at different weeks of exposure to the peanut containing CD. The first group (E1) received OVA to drink from day 7 through day 0 (week 1) of initiation of peanut containing CD. The second group (E2) received OVA to drink as of day 0 of the CD (week 1). The third group (E3), fourth (E4), and the fifth groups (E5) received OVA to drink as of the 7th, 14th, and 21st days after the beginning of the CD (weeks 2, 3, and 4), respectively (Table 1).
Bleeding Animals were bled 200 ml from the retrorbital plexus prior to manipulation, after each immunization and during the inflammation induction protocol. The serum was collected and stored at 20 1C until analyses.
Induction of the antigen-specific inflammatory gut reaction Two weeks after booster immunization with peanut protein extract, experimental animals were submitted to a CD with peanuts in natura, during a 30-day period as described by Teixeira (2003), Teixeira et al. (2008a) in press.
Determination of Ab levels An enzyme-linked immunosorbent assay (ELISA) was performed to detect total IgG. Briefly, 96-well micro plates (Alfa, Brazil) were coated with 10 mg of peanut protein extract or OVA in 0.1 M PBS buffer overnight at 4 1C per well. After a thorough wash the wells were blocked with PBS-gelatin before incubation of the serum samples for 3 h at RT. Detection was performed with goat anti-mouse IgG HRP (1:5000, Sigma) and OPD (o-phenylenediamine, Sigma). The reaction was interrupted with 0.1 M sulphuric acid and read at 492 nm in an ELISA micro plate reader (Anthos 2010, Germany). The results were expressed as arbitrary
Table 1. Timing of oral tolerance induction to Ovalbumin in the course of gut inflammation due to the exposure of a challenge diet containing the specific antigen to which animals had been previously immunized with peanut protein extract Challenge diet (weeks)
Group E1
1
1
OVAa
yb
Recovery (weeks) 2
3
4
1
2
3
4
5
6
7
yyc Group E2
OVA
y yy
Group E3
OVA
y yy
Group E4
OVA
y yy
Group E5
OVA
y yy
a
OVA 1 week of free access to sweetened egg white solution. y time of killing after induction of gut inflammation by the removal of normal mouse chow and introduction of challenge diet. c yy time of killing after recovery of gut inflammation by the removal of challenge diet and introduction of normal mouse chow. b
ARTICLE IN PRESS 272
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
units of ELISA corresponding to the area under the dilution curve of each serum.
12
Histomorphology As depicted in Table 1, animals of the experimental groups were euthanized at different times of CD exposure (1, 2, 3, and 4 weeks) and recovery period (1, 3, 4, 5, 6, and 7 weeks) to collect intestinal segments. Animals of the control groups were euthanized at the end of their respective experimental protocols diet), C1 – immune without gut inflammation (immunized with peanut protein without 4 weeks of CD); C2 – tolerant (OT to peanut protocol with 4 weeks of CD); C3 – immune (immunized with peanut protein with 4 weeks of CD); C4 – normal (sham immunized without exposure to CD). Intestinal segments of all animals were fixed with 10% buffered formaldehyde and stained with hematoxylineosin (HE) and analyzed to quantify gut alterations of the histological architecture. For the purpose of this study, we examined the histological parameters from the duodenum analyzing the general aspects of the tissue such as integrity of the intestinal structure, number of villi per field, edema, congestion and leukocyte infiltrate and establishing the ratios between villi height/width and intestinal epithelial cells/intraepithelial leukocytes (IEC/IEL).
Statistical analysis Statistical analysis was performed by using Fisher’s test, ANOVA, and Tukey0 s post test to determine the minimum significance difference (MSD) using GraphPad InStat program by GraphPad Software Incs.
Arbitrary Units of ELISA
10 Experimental 8
Tolerant Control
6 4 2 0 Zero
Booster
Challenge diet
Fig. 1. Total IgG anti-peanuts-Ab titers of tolerant, experimental and control animals before any manipulation (zero), after booster immunization (booster) and after CD (peanuts).
half of the animals of groups E1, E2, E3, and E4 was euthanized after 1, 2, 3, and 4 weeks of exposure, respectively. The rest of the animals were killed 1, 3, 4, 5, 6, and 7 weeks of recovery (removal of peanuts and reintroduction of normal mouse chow) as depicted in Table 1.
Macroscopic analysis of the intestine The macroscopic analysis of the peritoneum pertaining to animals of the experimental group showed a pale coloring of organs in general and revealed a frail consistency of the intestinal tissue as of the third week of CD in contrast to the control group. Animals of the recovery group also showed macroscopic improvement as of the third week (data not shown).
Microscopic analysis of the intestine
Results Immunogenicity and induction of OT to peanuts in natura Confirming prior data total protein peanut extract is immunogenic. All immunized animals responded with significantly higher levels of systemic antibodies (po0.001) when compared to saline controls. When animals had free access to the seeds in the diet for 7 consecutive days prior to immunization a significant reduction of specific Ab titers were observed compared to immunized group (po0.001) (Fig. 1).
Induction of antigen-specific inflammation of the gut To quantify the evolution of gut inflammation, induced by a peanut containing CD for a 4 week period
The first microscopic parameter of the duodenum evaluated was the villi height/villi width (vh/vw) ratio (Fig. 2). Immune animals of the group that had not received CD (C1) presented a normal vh/vw ratio (2.7970.63) as did the groups which received CD for one (E2) (2.0570.13) and 2 weeks (E3) (2.8370.75). As of the third week of CD significant alterations become evident. The most obvious architectural alterations were observed in the last experimental group E5 (1.3170.18), where significant shortening of the villi accompanied by enlargement of the vw was seen leading to a very significant difference (po0.01) when compared to the group without CD (C1) and tolerant control groups (C2). After removal of CD all groups presented recovery of the intestinal villi with significant changes as of the third week (Fig. 2). The second parameter evaluated was the number of the IEC in relation to the number of IEL per villi.
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Villi height Villi width
273
6
450 5 Villi height width ratio
Measure in µm
400 350 300 250 200 150 100
4 3 2 1
50 0 0
1
2
3
1
4 .
Time of challenge diet (weeks)
3
4
5
0
7
6
0
Time of recovery (weeks)
1
2
3
4
1 .
Time of challenge diet (weeks)
3
4
5
6
7
Time of recovery (weeks)
Fig. 2. Mean measures showing evolution of vh and vw (left panel) and the ratio established between these measures (right panel) during exposure to CD and recovery period of C57Bl/6J experimental mice (n ¼ 7).
300
IEC
10.0
IEL
9.0 8.0
200
7.0 IEC/IEL ratio
Number of cells per villi
250
150 100
6.0 5.0 4.0 3.0 2.0
50
1.0 0 0
1
2
3
4
Weeks of challenge diet
1
3
4
5
6
Weeks of recovery
7
0.0 0
1
2
3
4
Weeks of challenge diet
1
3
4
5
6
7
Weeks of recovery
Fig. 3. Mean intestinal epithelial cell (IEC) and intra epithelial leukocytes (IEL) count during induction of gut inflammation due to exposure to CD containing specific antigen and during recovery after reintroduction of conventional mouse chow (left panel). Establishment of the IEC/IEL ratio during CD and recovery period (right panel) (n ¼ 6).
The total number of cells decreases as the inflammatory process is installed. A positive correlation can be established between this cell count and histomorphology vh (data not shown). As the inflammatory process aggravates, during the 4 weeks of CD exposure, the IEC/IEL ratio decreases from 0.6070.07 (E2) to 0.3770.11 (E5). After the removal of the CD a normalization of the vh and cell count was seen, although these ratios do not reach normal levels for both parameters within the 7-week period observed. The improvement was slower for the cell count than for the architecture of the villi (Figs. 3 and 4). In Fig. 4, we demonstrate the differences between the different experimental groups. Representative examples of histological findings are shown in the microscopic analysis. In agreement with the macroscopy Fig. 4A shows a preserved intestinal structure of control animals
which had been immunized but had not received the CD. In Fig. 4B the mucosa of a tolerant animal demonstrates a similar architectural structure compared to the control group. On the other hand, figures C, D, E, and F, pertaining to groups E1–E4 killed after different times of CD exposure 1–4 weeks, respectively, demonstrates increasing alterations typical of chronic inflammatory process, such as: villi shortening, villi widening, edema, congestion, and leukocyte infiltration in duodenum.
Timing of OVA feeding in mice with antigen-specific inflammation of the gut The experimental groups (E1–E5) that were submitted to a diet containing sweetened OVA ad libitum during 7
ARTICLE IN PRESS 274
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Fig. 4. Histomorphology (HE) of small intestine – (A) Control animal. (B) Tolerant animal. (C) Animal with 1 week of gut inflammation induction. (D) Animal with 2 weeks of gut inflammation induction. (E) Animal with 3 weeks of gut inflammation induction. (F) Animal with 4 weeks of gut inflammation induction. Progressive alterations such as presence of edema, leukocyte infiltrate, modification of normal architecture of villi can be observed in panels C, D, E, and F.
8.00 Arbitrary Units of ELISA
7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 E1
E2
E3
E4
Experimental groups
E5
Tolerant
Immune
Control groups
Fig. 5. (}) Individual and (–) mean anti-OVA IgG, after free ingestion of sweetened OVA in the diet, prior (E1) and during (E2–E5) the induction of an antigen-specific gut inflammation due to a peanut containing CD, compared to normal and immune control groups (n ¼ 7).
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
275
8.00 Arbitrary Units of ELISA
7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 E1
E2
E3
E4
E5
Tolerant
Experimental groups
Immune
Control groups
Fig. 6. (}) Individual and (–) mean anti-OVA IgG, after OVA booster in groups that had free access to sweetened OVA in the diet, prior (E1) and during (E2–E5) the induction of an antigen-specific gut inflammation due to a peanut containing CD. (n ¼ 7) compared to normal and immune control groups.
Arbitrary Units Of ELISA
6 5
E1 E2
4
E3 E4
3
E5
2
Immune
Tolerant
1 0 Oral exposure
Booster
Experimental Groups
Primary
Booster
Control Groups
Fig. 7. Comparison of mean anti-OVA IgG titers of control and experimental groups. Experimental groups: anti-OVA Ab titers after Ova feeding over a 7-day period – (E1) began 7 days prior to peanut containing CD (E2) on day 0, (E3) on day 7, (E4) on day 14 and (E5) on day 21 of CD and after the booster inoculation of OVA. Control groups: Ab titers after primary and booster inoculation.
days at different timings of the peanut containing CD presented increasing anti-OVA specific Ab titers. The anti-OVA Ab titers after oral exposure to sweetened egg white and prior to systemic immunization of the 5 groups showed that animals of E1 (0.5770.14), E2 (1.0670.41) and E3 (0.7170.27) present low mean Ab titers, respectively) not significantly different from tolerant controls. Groups E4 (1.5970.49) and E5 (2.5471.07) present anti-OVA Ab titers compatible to primary systemic immunization and are significantly different from groups E1–E3 (po0.01) (Fig. 5). Although all groups presented a significant elevation of total Ab titers after systemic immunization animals of group E1 (1.3270.468) presented the lowest elevation. The Ab titers of group E1 differ significantly from
immune control animals (po0.001) but not from tolerant animals (p40.5) thus being considered tolerant. These animals also presented a significant difference from all other experimental groups. On the other hand groups E2 (4.1170.62), E3 (4.3570.38), E4 (3.3670.28), and E5 (5.4770.73) present significant differences when compared to tolerant group but not to Immunized group thus being considered systemically sensitized. When comparing these latter groups (E2–E5) a scatter in the individual Ab titers can be seen (Fig. 6). Fig. 7 depicts the mean elevation of Ab titers between oral administration of OVA and systemic immunization indicating that all animals that receive OVA to drink as of the introduction of the peanut containing CD become systemically sensitized to OVA.
ARTICLE IN PRESS 276
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Discussion Food allergy is a disease that affects approximately 6% of young children and 3.7% of the adult population in the USA (Scott and Sampson, 2006). Recognition of food allergy is largely based on symptoms reported by patients or parents. While IgE-mediated food allergy is relatively easy to diagnose, identification of non-IgEmediated food allergy relies on suggestive symptoms, frequently mimicked by various other adverse reactions to foods and other diseases. (Engelmann et al., 2008). Systemic introduction of an antigen whether by injection (subcutaneous or intramuscular) or injury, may lead to local infiltration of inflammatory cells and specific immunoglobulin production (Strid et al., 2005). In contrast, antigens introduced at mucosal surfaces generally induce active immune tolerance to the antigens (Mayer and Shao, 2004). In this work, we confirm that peanut antigens elicit qualitatively distinct immune responses based on their initial rout of entry and on prior immune experience of the animal to the respective antigen. As we have previously described when mice and rats are systemically immunized to peanut protein extract and then exposed exclusively to a diet based on raw peanuts they develop inflammation of the gut. The features of this model are prominent mononuclear leukocyte infiltrate and lamina propria edema and larger numbers of goblets cells inducing architectural changes such as, villous atrophy and crypt hyperplasia. Clinically, these animals present significant weight loss which can be explained by the histological findings of the gut, which in turn may be the cause of lower food intake (Teixeira et al., 2008b), and a lower absorption rate (Teixeira et al., 2008a, in press). These are in agreement to the typical signs seen in human celiac disease (gluten enteropathy), and which are observed in other foodinduced enteropathies. We have also described that when immunized animals are given the choice, they avoid the allergenic foodstuff eating the other components of the diet and consequently do not develop gut alterations (Teixeira et al., 2008b). Foods contain substances that can both control the physiological functions of the body and modulate the immune responses (Kaminogawa and Nanno, 2004). Although both OT and allergic responses have been established both in humans and in animal models these depend on complex interactions involving the genetic background, the type, timing and dose of Ag and is thought to be a consequence of the introduction of proteins in the gut in physiologic conditions (Vaz and Pordeus, 2005; Faria and Weiner, 2005; Teixeira et al., 2008a, b in press). Furthermore, dealing specifically with it has been shown that the allergic reactions to foods may provoke characteristic responses in the skin, gastrointestinal and
respiratory tract, regardless of the immunopathogenic mechanism responsible for the reaction (Sampsom et al., 2005). As such allergic responses require complex interactions between the protein and the immune system, which are notoriously difficult to predict (Huby et al., 2000). There has been considerable recent broadening of basic concepts of intestinal food allergy, in particular the importance of non-IgE-mediated mechanisms. The traditional emphasis on IgE-mediated allergy now appears inappropriate in light of current studies of the basic mechanisms of OT to dietary antigen and of increasing recognition of the requirement for early infectious challenge in the prevention of allergic sensitization (Chehade, 2007). This major change in emphasis has been elicited both by basic scientific studies and by the recognition of novel patterns of food allergic diseases within the pediatric population. A rapid increase in food-allergic sensitization has been noted in the last decade in special, two previously rare phenomena: multiple food allergies and sensitization of exclusively breast-fed infants to antigens eaten by the mother have become commonplace (Murch, 2000). In our study, we propose that prior systemic immunization associated to the port of entry of the specific antigen is a strong cause of induction of chronic inflammatory reactions of the gut. Although other sensitizing mechanisms that simulate the clinical setting such as skin sensitization (Strid et al., 2005) or oral sensitization associated to cholera toxin (Mayer, 2005) could have been used, we chose to sensitize animals using the subcutaneous rout in order to quantify the exact amount of antigen that was given. In our results, the group of animals which received OVA before (E1) the initiation of the 4-week peanut CD presented specific Ab titers similar to the tolerant group. Animals which received free access to OVA during the first and second week of the CD (E2 and E3) presented specific Ab titers similar to the primary immunization of immune control animals. As of the third week of CD, we show that the specific Ab titers of the groups that received OVA during this period, groups E4 and E5, were similar booster level of the immune control group. After booster immunization group E1 continued to present low Ab titers in contrast to all other experimental groups (E2–E5) that received OVA to drink as of day 1 of CD exposure. These latter groups presented Ab titers equivalent or higher to immune control group confirming their immunological status. Thus, oral administration of OVA in mice sensitized to peanuts before the induction of chronic inflammation due to the exposure to a CD containing the specific antigen resulted in the development of OT. In contrast, as of the exposure to the CD which induces significant inflammatory changes to the gut, the introduction of the
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
novel protein led to a shift from the induction of OT to systemic immunization. Although we have not determined the profile of the cells that infiltrate the gut during CD we could argue that these are cells with an inflammatory profile which have been elicited during systemic immunization and that migrate to the gut upon exposure to the specific antigen inducing a shift from a tolerogenic to an immunizing environment in the gut. In contrast to current thought that peanut allergy is primarily a Th2-driven process, there is accumulating data from the literature indicating that this food allergy may be a result of a mixed Th1Th2 immune response. The absence of systemic anaphylactic reactions, the presence of a mononuclear infiltrate with few mast cells and eosinophils in our murine model corroborates to the idea that this is a mixed Th1Th2 modulation of peanut allergy. Finally, the verification of similar inflammatory responses in the gut of mice with different genetic backgrounds (Balb/c and C57BL/B6) provides an interesting method for the study of this mixed immune response to peanut proteins. The animals that were submitted to a chronic gut inflammatory process showed more heterogeneity in the individual immune responses when compared to systemic immunization. This may be due to the fact that the animals were free to determine the amount of daily intake. The intrinsic potential of protein allergens to induce sensitization will be manifest only in susceptible individuals, and then only if the allergen is encountered in sufficient quantities and via a relevant route of exposure (Huby et al., 2000). In our model this fact can be seen in the animals that are immunized (thus susceptible) but not challenged in the diet in which the introduction of a novel protein leads to OT due to normal gut environment. Preliminary results indicate that during the recovery period of the chronic inflammation of the gut a concomitant recovery of the ability to induce OT is shown arguing that the deregulation of the normal physiology is transitory depending on local exposure to the allergen. In conclusion the induction of OT or systemic immunization to a novel protein depends on the local environment of the gut. Taking these results to the clinical setting we could suggest that children with gut alterations should not be exposed to novel proteins as a measure to avoid multiple food allergies.
References Bruijnzeel-Koomen, C., Ortolani, C., Aas, K., BindslevJensen, C., Bjo¨rkste´n, B., Moneret-Vautrin, D., Wu¨thrich, B., 1995. Adverse reactions to food. European academy of allergology and clinical immunology subcommittee. Allergy 50, 623–635.
277
Chehade, M., 2007. IgE and non-IgE-mediated food allergy: treatment in 2007. Curr. Opin. Allergy Clin. Immunol. 7, 264–268. Engelmann, P.A., Beyer, K., Wesley Burks, A., Lack, G., Liacouras, C.A., Hourihane, J.O., Sampson, H.A., Sodergren, E., 2008. New visions for food allergy: an iPAC summary and future trends. Pediatr. Allergy Immunol. (Suppl. 19), 26–39. Faria, A., Weiner, H., 2005. Oral tolerance. Immunol. Rev. 206, 232–259. Huby, R., Dearman, R., Kimber, I., 2000. Why are some proteins allergens? Toxicol. Sci. 55, 235–246. Husby, S., Mestecky, J., Moldoveanu, Z., Holland, S., Elson, C., 1994. Oral tolerance in humans. T cell but not B cell tolerance after antigen feeding. J. Immunol. 152, 4663–4670. Johansson, S.G., Hourihane, J.O., Bousquet, J., BruijnzeelKoomen, C., Dreborg, S., Haahtela, T., Kowalski, M.L., Mygind, N., Ring, J., van Cauwenberge, P., van HageHamsten, M., Wu¨thrich, B., 2001. EAACI (the European academy of allergology and cinical immunology) nomenclature task force. A revised nomenclature for allergy. An EAACI position statement from the EAACI nomenclature task force. Allergy 56, 813–824. Kaminogawa, S., Nanno, M., 2004. Modulation of immune functions by foods. Evid-Based Complement. Altern. Med. 1, 241–250. Kraus, T.A., Toy, L., Chan, L., Childs, J., Cheifetz, A., Mayer, L., 2004a. Failure to induce oral tolerance in Cronh’s and ulcerative colitis patients: possible genetic risk. Ann. NY Acad. Sci. 20, 225–238. Kraus, TA., Toy, L., Chan, L., Childs, J., Mayer, L., 2004b. Failure to induce oral tolerance to a soluble protein in patients with inflammatory bowel disease. Gastoenterology 126, 1771–1778. Kuchroo, V., Das, M., Brown, J., Ranger, A., 1995. B7-1 and B7-2 costimulatory molecules activate differentially the Th1/Th2 developmental pathways: application to autoimmune disease therapy. Cell 80, 707–718. Mayer, L., 2005. Mucosal immunity. Immunol. Rev. 206, 5–21. Mayer, L., Shao, L., 2004. Therapeutic potencial of oral tolerance. Nat. Rev. 4, 1–14. Murch, S., 2000. The immunologic basis for intestinal food allergy. Curr. Opin. Gastroenterol. 16, 552–557. Neurath, M., Finotto, S., Glimcher, L., 2002. The role of Th1/Th2 polarization in mucosal immunity. Nat. Med. 8, 567–573. Nowak-Wegrzyn, A., 2003. Future approaches to food allergy. Pediatrics 111, 1672–1680. Ortolani, C., Bruijnzeel-Koomen, C., Bengtsson, U., BindslevJensen, C., Bjo¨rkste´n, B., Høst, A., Ispano, M., Jarish, R., Madsen, C., Nekam, K., Paganelli, R., Poulsen, L.K., Wu¨thrich, B., 1999. Controversial aspects of adverse reactions to food. European academy of allergology and clinical immunology (EAACI) reactions to food subcommittee. Allergy 54, 27–45. Sampsom, J., Van Berkel, L., Van Helvoort, J., Unger, W., Jansen, W., Thepen, T., Mebius, R., Verbeek, S., Kraal, G., 2005. FcgRIIB regulates nasal and oral tolerance: a role for dendritic cells. J. Immunol. 174, 5279–5287.
ARTICLE IN PRESS 278
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Scott, H., Sampson, H., 2006. Food allergy. J. Allergy Clin. Immunol. 117, 470–475. Strid, J., Hourihane, J., Kimber, I., Callard, R., Strobel, S., 2005. Epicutaneous exposure to peanut protein prevents oral tolerance and enhances allergic sensitization. Clin. Exp. Allergy 35, 757–766. Teixeira, G., 2003. Um Modelo Murino de Inflamac¸a˜o Intestinal Croˆnica Toleraˆncia e Imunizac¸a˜o Oral: antigenicidade versus Temporalidade. Universidade Federal Fluminense, Nitero´i, p. 213. Teixeira, G., Antunes, D., Castro Jr., A., Campos, S., Pedruzzi, M., Paschoal, P., Costa, J., Arau´jo, V., Andrade, L., Moreno, S., Caldas, L., Cardoso, G., Nobrega, A., 2008a. Induction of an antigen specific gut
inflammatory reaction in mice and rats: a model for human inflammatory bowel disease. Braz. Arch. Biol. Technol. (in press). Teixeira, G., Paschoal, P., Oliveira, V., Pedruzzi, M., Campos, S., Andrade, L., No´brega, A., 2008b. Diet selection in immunologically manipulated mice. Immunobiology 213, 1–12. Vaz, N., Pordeus, V., 2005. Visita a` imunologia. Arq. Bras. Cardiol. 85, 350–362. Worbs, T., Bode, U., Yan, S., Hoffmann, M., Hintzen, G., Bernhardt, G., Foster, R., Pabst, O., 2006. Oral tolerance originates in the intestinal immune system and relies on antigen carriage by dendritic cells. J. Exp. Med. 203, 519–527.
Immunobiology 214 (2009) 269–278 www.elsevier.de/imbio
Food allergy/hypersensitivity: Antigenicity or timing? Patrı´cia Olaya Paschoala,b,, Sylvia M.N. Camposb, Monique M.B. Pedruzzib, Vale´ria Garridob, Moˆnica Bissoa, Danielle M.F. Antunesb, Alberto F. Nobregac, Gerlinde Teixeirab a
Programa de Po´s-Graduac¸a˜o em Patologia, Faculdade de Medicina, Universidade Federal Fluminense, Brazil Laborato´rio do Grupo de Imunologia Gastrintestinal, Departamento de Imunobiologia, Instituto de Biologia, Universidade Federal Fluminense, Brazil c Universidade Federal do Rio de Janeiro b
Received 30 October 2007; received in revised form 25 August 2008; accepted 15 September 2008
Abstract Mechanisms involved in the induction of oral tolerance (OT) or systemic immunization through the oral rout are still poorly understood. In our previous studies, we have shown that when normal mice eat peanuts they become tolerant, with no gut alterations. Conversely, if immunized with peanut proteins prior to a challenge diet (CD) containing peanuts they develop chronic inflammation of the gut. Our aim is to evaluate the consequences of the introduction of a novel protein in the diet of animals presenting antigen-specific gut inflammation. Adult, female C57BL/6J mice were divided in control (C) and experimental (E) groups. C1–C3 received peanut protein immunization, animals of the control groups C4 were sham immunized, and control group C5 received ovalbumin (OVA) immunization. The experimental group was immunized with peanut protein extract. Before initial exposure to a 30-day peanut containing CD, the experimental group was divided into 5 groups (E1–E5). OVA feeding began 7 days prior CD (E1) on day 0 (E2), 7 (E3), 14 (E4) and 21 (E5) during CD. Our results show that oral exposure to a novel protein (OVA) in the absence of gut inflammation (E1) leads to low levels of systemic antibody (Ab) titers, comparable to tolerant animals. Conversely, as off initial induction of inflammation, groups submitted to OVA (OT) protocol develop increasingly higher systemic Ab titers similar to animals of the immune control group. In conclusion, our protocol indicates that timing is more important than the antigenicity when a novel protein is offered, in the diet. r 2008 Elsevier GmbH. All rights reserved. Keywords: Food allergy; Ovalbumin; Oral tolerance; Peanuts
Introduction For many years the term allergy was used as a synonym for IgE-mediated hypersensitivity, thus Corresponding author at: Instituto de Biologia, Alameda Barros Terra s/n, Centro, Nitero´i, 24020-150, Rio de Janeiro, Brazil. Tel.: 55 21 26292313. E-mail address: [email protected] (P.O. Paschoal).
0171-2985/$ - see front matter r 2008 Elsevier GmbH. All rights reserved. doi:10.1016/j.imbio.2008.09.007
associated to severe potentially life-threatening anaphylactic reactions. However, clinical observations have shown that not all allergies have this as the underlining mechanism. A force task of the European Academy of Allergy and Clinical Immunology published in 2001 a revised nomenclature for allergies (Johansson et al., 2001). This force task proposed that allergies are hypersensitivity reactions initiated by immunologic mechanisms and that adverse reactions to food should
ARTICLE IN PRESS 270
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
be called food hypersensitivity. When immunologic mechanisms have been demonstrated, the appropriate term is food allergy, and, if the role of IgE is highlighted, the term is IgE-mediated food allergy. All other reactions, previously referred to as ‘‘food intolerance’’, should be referred to as nonallergic food hypersensitivity as previously suggested by Bruijnzeel-Koomen et al. (1995) and Ortolani et al. (1999). In 1946, Merill Chase showed that oral administration of a contact-sensitizing agent (2,4-dinitrochlorobenzene) does not lead to sensitization, but rather prevents the animal from eliciting an immune response upon subsequent subcutaneous injections or cutaneous challenge. The specific induction of these regulated responses by administration of antigen through the gastrointestinal tract is known as oral tolerance (OT) (Mayer and Shao, 2004). Thus, food allergy represents an abrogation of normal OT to that specific foodstuff (Scott and Sampson, 2006). For efficient strategies aimed at an active cure of food allergies, a better understanding of the pathogenic mechanisms is strongly needed (Sampsom et al., 2005). In contrast to current thought that food allergy is primarily a Th2-driven process, there is accumulating data from the literature indicating that food allergies may also be a result of a mixed Th1Th2 immune response. The mucosal immune system utilizes a very complex and tightly regulated system of suppression and controlled inflammation to distinguish between innocuous and potentially harmful antigens of the intestinal lumen. Although the exact mechanisms governing this regulation remains to be elucidated, evidence exists for the involvement of a variety of cell types, including CD4+, CD8+ T cell, regulatory T cells, B cells, and dendritic cells (Mayer, 2005). When both the ability to remain tolerant to the abundance of dietary antigens and prevention of luminal bacterial flora to harm the host fail, disease may install (Kraus et al., 2004a). OT induction is a key feature of intestinal immunity, generating systemic low responsiveness to ingested antigens (Worbs et al., 2006). Mechanisms governing tolerance induction have been well characterized in selected mouse strains. Similar studies in man are lacking, although there is evidence that tolerance can be induced. The ability to induce OT in humans was reported by Husby et al., although the type of tolerance generated to a neo-antigen, keyhole limpet hemocyanin (KLH), was slightly different from that seen in murine models. Tolerance for delayed-type hypersensitivity (DTH) responses but not antibody (Ab) responses was seen (Husby et al., 1994; Kraus et al., 2004b). In disease states, tolerance is altered and this may account for the presence of mucosal inflammation. In food hypersensitivity there is evidence that allergens may be handled differently and this may play a role
in disease expression (Mayer and Shao, 2004). It has been clearly demonstrated that an upregulated Th1 immune response and subsequent overproduction of Th1 cytokines may lead to autoimmune diseases such as, inflammatory bowel disease, autoimmune arthritis, experimental autoimmune encephalomyelitis, and typeI diabetes (Kuchroo et al., 1995; Neurath et al., 2002). Thus, there is an intense interest in understanding the regulation of the immune response, with the goal of developing therapeutic strategies. In cases of multiple food allergies, restricted diets may result in unbalanced nutrition and pose a considerable hardship to the family of a food-allergic child (Nowak-Wegrzyn, 2003). In previous studies of our group, we have shown that crude peanut protein extract is immunogenic both in mice and rats. All animals responded with high levels of systemic antibodies (po0.001) as compared to saline controls. No differences were observed when comparing animals within strains nor when comparing young adults (2–4 months) with older animals (12–16 months). Submitting animals to a peanut containing challenge diet (CD) induces important alterations of the gut mucosal architecture, typical of chronic inflammation in immunized but not in normal animals (Teixeira, 2003; Teixeira et al., 2008a, in press). In this study, we show the consequences of the administration of a non related protein in the presence of a chronic antigen-specific gut inflammation.
Methods Animals C57BL/6J female mice, 2–3 months old, were obtained by the local Animal Resources Center (Universidade Federal Fluminense). These animals were randomly divided into 2 groups: control and experimental. These were subsequently subdivided into three control groups (n ¼ 7) and 5 experimental groups (n ¼ 14). All animals were individually marked, to be able to perform paired statistical analysis. This research was approved by the animal ethics committee of the School of Medicine of the Fluminense Federal University.
Immunization protocols The immunization protocol utilized in this study was: 100 mg specific protein (peanut protein extract or ovalbumin (OVA)) plus 1 mg of adjuvant [Al(OH)3], in a final volume of 200 ml by subcutaneous route (sc) for the primary immunization and 100 mg of specific protein without adjuvant, performed 21 days later, for the booster immunization.
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
For normal controls the animals were sham immunized with 1 mg of adjuvant [Al(OH)3], in a final volume of 200 ml of saline by sc for the primary immunization and saline without adjuvant, performed 21 days later, for the booster immunization. All animals of the experimental groups (E1–E5) and control groups C1–C3 received peanut protein immunization, animals of the control groups C4 were sham immunized, and control group C5 received OVA immunization.
Tolerization to OVA or peanuts To induce OT, animals received to drink a sweetened 20% egg white solution (diluted in saline v/v with 5% of saccharine) or to eat a diet containing exclusively peanut seeds for 7 days ad libitum prior to the immunization protocol described above with the respective proteins. (Teixeira, 2003; Teixeira et al., 2008a, in press).
271
Animals of control group immunized to peanuts (C1) continued to receive commercial chow, (C2) which had been submitted to peanuts OT protocol and (C3) immunized to peanuts received peanuts for a 30-day period.
Timing of OVA feeding in mice with antigen-specific inflammation of the gut To determine the consequences of gut inflammation on the induction of OT, the 5 experimental animals groups received OVA to drink at different weeks of exposure to the peanut containing CD. The first group (E1) received OVA to drink from day 7 through day 0 (week 1) of initiation of peanut containing CD. The second group (E2) received OVA to drink as of day 0 of the CD (week 1). The third group (E3), fourth (E4), and the fifth groups (E5) received OVA to drink as of the 7th, 14th, and 21st days after the beginning of the CD (weeks 2, 3, and 4), respectively (Table 1).
Bleeding Animals were bled 200 ml from the retrorbital plexus prior to manipulation, after each immunization and during the inflammation induction protocol. The serum was collected and stored at 20 1C until analyses.
Induction of the antigen-specific inflammatory gut reaction Two weeks after booster immunization with peanut protein extract, experimental animals were submitted to a CD with peanuts in natura, during a 30-day period as described by Teixeira (2003), Teixeira et al. (2008a) in press.
Determination of Ab levels An enzyme-linked immunosorbent assay (ELISA) was performed to detect total IgG. Briefly, 96-well micro plates (Alfa, Brazil) were coated with 10 mg of peanut protein extract or OVA in 0.1 M PBS buffer overnight at 4 1C per well. After a thorough wash the wells were blocked with PBS-gelatin before incubation of the serum samples for 3 h at RT. Detection was performed with goat anti-mouse IgG HRP (1:5000, Sigma) and OPD (o-phenylenediamine, Sigma). The reaction was interrupted with 0.1 M sulphuric acid and read at 492 nm in an ELISA micro plate reader (Anthos 2010, Germany). The results were expressed as arbitrary
Table 1. Timing of oral tolerance induction to Ovalbumin in the course of gut inflammation due to the exposure of a challenge diet containing the specific antigen to which animals had been previously immunized with peanut protein extract Challenge diet (weeks)
Group E1
1
1
OVAa
yb
Recovery (weeks) 2
3
4
1
2
3
4
5
6
7
yyc Group E2
OVA
y yy
Group E3
OVA
y yy
Group E4
OVA
y yy
Group E5
OVA
y yy
a
OVA 1 week of free access to sweetened egg white solution. y time of killing after induction of gut inflammation by the removal of normal mouse chow and introduction of challenge diet. c yy time of killing after recovery of gut inflammation by the removal of challenge diet and introduction of normal mouse chow. b
ARTICLE IN PRESS 272
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
units of ELISA corresponding to the area under the dilution curve of each serum.
12
Histomorphology As depicted in Table 1, animals of the experimental groups were euthanized at different times of CD exposure (1, 2, 3, and 4 weeks) and recovery period (1, 3, 4, 5, 6, and 7 weeks) to collect intestinal segments. Animals of the control groups were euthanized at the end of their respective experimental protocols diet), C1 – immune without gut inflammation (immunized with peanut protein without 4 weeks of CD); C2 – tolerant (OT to peanut protocol with 4 weeks of CD); C3 – immune (immunized with peanut protein with 4 weeks of CD); C4 – normal (sham immunized without exposure to CD). Intestinal segments of all animals were fixed with 10% buffered formaldehyde and stained with hematoxylineosin (HE) and analyzed to quantify gut alterations of the histological architecture. For the purpose of this study, we examined the histological parameters from the duodenum analyzing the general aspects of the tissue such as integrity of the intestinal structure, number of villi per field, edema, congestion and leukocyte infiltrate and establishing the ratios between villi height/width and intestinal epithelial cells/intraepithelial leukocytes (IEC/IEL).
Statistical analysis Statistical analysis was performed by using Fisher’s test, ANOVA, and Tukey0 s post test to determine the minimum significance difference (MSD) using GraphPad InStat program by GraphPad Software Incs.
Arbitrary Units of ELISA
10 Experimental 8
Tolerant Control
6 4 2 0 Zero
Booster
Challenge diet
Fig. 1. Total IgG anti-peanuts-Ab titers of tolerant, experimental and control animals before any manipulation (zero), after booster immunization (booster) and after CD (peanuts).
half of the animals of groups E1, E2, E3, and E4 was euthanized after 1, 2, 3, and 4 weeks of exposure, respectively. The rest of the animals were killed 1, 3, 4, 5, 6, and 7 weeks of recovery (removal of peanuts and reintroduction of normal mouse chow) as depicted in Table 1.
Macroscopic analysis of the intestine The macroscopic analysis of the peritoneum pertaining to animals of the experimental group showed a pale coloring of organs in general and revealed a frail consistency of the intestinal tissue as of the third week of CD in contrast to the control group. Animals of the recovery group also showed macroscopic improvement as of the third week (data not shown).
Microscopic analysis of the intestine
Results Immunogenicity and induction of OT to peanuts in natura Confirming prior data total protein peanut extract is immunogenic. All immunized animals responded with significantly higher levels of systemic antibodies (po0.001) when compared to saline controls. When animals had free access to the seeds in the diet for 7 consecutive days prior to immunization a significant reduction of specific Ab titers were observed compared to immunized group (po0.001) (Fig. 1).
Induction of antigen-specific inflammation of the gut To quantify the evolution of gut inflammation, induced by a peanut containing CD for a 4 week period
The first microscopic parameter of the duodenum evaluated was the villi height/villi width (vh/vw) ratio (Fig. 2). Immune animals of the group that had not received CD (C1) presented a normal vh/vw ratio (2.7970.63) as did the groups which received CD for one (E2) (2.0570.13) and 2 weeks (E3) (2.8370.75). As of the third week of CD significant alterations become evident. The most obvious architectural alterations were observed in the last experimental group E5 (1.3170.18), where significant shortening of the villi accompanied by enlargement of the vw was seen leading to a very significant difference (po0.01) when compared to the group without CD (C1) and tolerant control groups (C2). After removal of CD all groups presented recovery of the intestinal villi with significant changes as of the third week (Fig. 2). The second parameter evaluated was the number of the IEC in relation to the number of IEL per villi.
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Villi height Villi width
273
6
450 5 Villi height width ratio
Measure in µm
400 350 300 250 200 150 100
4 3 2 1
50 0 0
1
2
3
1
4 .
Time of challenge diet (weeks)
3
4
5
0
7
6
0
Time of recovery (weeks)
1
2
3
4
1 .
Time of challenge diet (weeks)
3
4
5
6
7
Time of recovery (weeks)
Fig. 2. Mean measures showing evolution of vh and vw (left panel) and the ratio established between these measures (right panel) during exposure to CD and recovery period of C57Bl/6J experimental mice (n ¼ 7).
300
IEC
10.0
IEL
9.0 8.0
200
7.0 IEC/IEL ratio
Number of cells per villi
250
150 100
6.0 5.0 4.0 3.0 2.0
50
1.0 0 0
1
2
3
4
Weeks of challenge diet
1
3
4
5
6
Weeks of recovery
7
0.0 0
1
2
3
4
Weeks of challenge diet
1
3
4
5
6
7
Weeks of recovery
Fig. 3. Mean intestinal epithelial cell (IEC) and intra epithelial leukocytes (IEL) count during induction of gut inflammation due to exposure to CD containing specific antigen and during recovery after reintroduction of conventional mouse chow (left panel). Establishment of the IEC/IEL ratio during CD and recovery period (right panel) (n ¼ 6).
The total number of cells decreases as the inflammatory process is installed. A positive correlation can be established between this cell count and histomorphology vh (data not shown). As the inflammatory process aggravates, during the 4 weeks of CD exposure, the IEC/IEL ratio decreases from 0.6070.07 (E2) to 0.3770.11 (E5). After the removal of the CD a normalization of the vh and cell count was seen, although these ratios do not reach normal levels for both parameters within the 7-week period observed. The improvement was slower for the cell count than for the architecture of the villi (Figs. 3 and 4). In Fig. 4, we demonstrate the differences between the different experimental groups. Representative examples of histological findings are shown in the microscopic analysis. In agreement with the macroscopy Fig. 4A shows a preserved intestinal structure of control animals
which had been immunized but had not received the CD. In Fig. 4B the mucosa of a tolerant animal demonstrates a similar architectural structure compared to the control group. On the other hand, figures C, D, E, and F, pertaining to groups E1–E4 killed after different times of CD exposure 1–4 weeks, respectively, demonstrates increasing alterations typical of chronic inflammatory process, such as: villi shortening, villi widening, edema, congestion, and leukocyte infiltration in duodenum.
Timing of OVA feeding in mice with antigen-specific inflammation of the gut The experimental groups (E1–E5) that were submitted to a diet containing sweetened OVA ad libitum during 7
ARTICLE IN PRESS 274
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Fig. 4. Histomorphology (HE) of small intestine – (A) Control animal. (B) Tolerant animal. (C) Animal with 1 week of gut inflammation induction. (D) Animal with 2 weeks of gut inflammation induction. (E) Animal with 3 weeks of gut inflammation induction. (F) Animal with 4 weeks of gut inflammation induction. Progressive alterations such as presence of edema, leukocyte infiltrate, modification of normal architecture of villi can be observed in panels C, D, E, and F.
8.00 Arbitrary Units of ELISA
7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 E1
E2
E3
E4
Experimental groups
E5
Tolerant
Immune
Control groups
Fig. 5. (}) Individual and (–) mean anti-OVA IgG, after free ingestion of sweetened OVA in the diet, prior (E1) and during (E2–E5) the induction of an antigen-specific gut inflammation due to a peanut containing CD, compared to normal and immune control groups (n ¼ 7).
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
275
8.00 Arbitrary Units of ELISA
7.00 6.00 5.00 4.00 3.00 2.00 1.00 0.00 E1
E2
E3
E4
E5
Tolerant
Experimental groups
Immune
Control groups
Fig. 6. (}) Individual and (–) mean anti-OVA IgG, after OVA booster in groups that had free access to sweetened OVA in the diet, prior (E1) and during (E2–E5) the induction of an antigen-specific gut inflammation due to a peanut containing CD. (n ¼ 7) compared to normal and immune control groups.
Arbitrary Units Of ELISA
6 5
E1 E2
4
E3 E4
3
E5
2
Immune
Tolerant
1 0 Oral exposure
Booster
Experimental Groups
Primary
Booster
Control Groups
Fig. 7. Comparison of mean anti-OVA IgG titers of control and experimental groups. Experimental groups: anti-OVA Ab titers after Ova feeding over a 7-day period – (E1) began 7 days prior to peanut containing CD (E2) on day 0, (E3) on day 7, (E4) on day 14 and (E5) on day 21 of CD and after the booster inoculation of OVA. Control groups: Ab titers after primary and booster inoculation.
days at different timings of the peanut containing CD presented increasing anti-OVA specific Ab titers. The anti-OVA Ab titers after oral exposure to sweetened egg white and prior to systemic immunization of the 5 groups showed that animals of E1 (0.5770.14), E2 (1.0670.41) and E3 (0.7170.27) present low mean Ab titers, respectively) not significantly different from tolerant controls. Groups E4 (1.5970.49) and E5 (2.5471.07) present anti-OVA Ab titers compatible to primary systemic immunization and are significantly different from groups E1–E3 (po0.01) (Fig. 5). Although all groups presented a significant elevation of total Ab titers after systemic immunization animals of group E1 (1.3270.468) presented the lowest elevation. The Ab titers of group E1 differ significantly from
immune control animals (po0.001) but not from tolerant animals (p40.5) thus being considered tolerant. These animals also presented a significant difference from all other experimental groups. On the other hand groups E2 (4.1170.62), E3 (4.3570.38), E4 (3.3670.28), and E5 (5.4770.73) present significant differences when compared to tolerant group but not to Immunized group thus being considered systemically sensitized. When comparing these latter groups (E2–E5) a scatter in the individual Ab titers can be seen (Fig. 6). Fig. 7 depicts the mean elevation of Ab titers between oral administration of OVA and systemic immunization indicating that all animals that receive OVA to drink as of the introduction of the peanut containing CD become systemically sensitized to OVA.
ARTICLE IN PRESS 276
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Discussion Food allergy is a disease that affects approximately 6% of young children and 3.7% of the adult population in the USA (Scott and Sampson, 2006). Recognition of food allergy is largely based on symptoms reported by patients or parents. While IgE-mediated food allergy is relatively easy to diagnose, identification of non-IgEmediated food allergy relies on suggestive symptoms, frequently mimicked by various other adverse reactions to foods and other diseases. (Engelmann et al., 2008). Systemic introduction of an antigen whether by injection (subcutaneous or intramuscular) or injury, may lead to local infiltration of inflammatory cells and specific immunoglobulin production (Strid et al., 2005). In contrast, antigens introduced at mucosal surfaces generally induce active immune tolerance to the antigens (Mayer and Shao, 2004). In this work, we confirm that peanut antigens elicit qualitatively distinct immune responses based on their initial rout of entry and on prior immune experience of the animal to the respective antigen. As we have previously described when mice and rats are systemically immunized to peanut protein extract and then exposed exclusively to a diet based on raw peanuts they develop inflammation of the gut. The features of this model are prominent mononuclear leukocyte infiltrate and lamina propria edema and larger numbers of goblets cells inducing architectural changes such as, villous atrophy and crypt hyperplasia. Clinically, these animals present significant weight loss which can be explained by the histological findings of the gut, which in turn may be the cause of lower food intake (Teixeira et al., 2008b), and a lower absorption rate (Teixeira et al., 2008a, in press). These are in agreement to the typical signs seen in human celiac disease (gluten enteropathy), and which are observed in other foodinduced enteropathies. We have also described that when immunized animals are given the choice, they avoid the allergenic foodstuff eating the other components of the diet and consequently do not develop gut alterations (Teixeira et al., 2008b). Foods contain substances that can both control the physiological functions of the body and modulate the immune responses (Kaminogawa and Nanno, 2004). Although both OT and allergic responses have been established both in humans and in animal models these depend on complex interactions involving the genetic background, the type, timing and dose of Ag and is thought to be a consequence of the introduction of proteins in the gut in physiologic conditions (Vaz and Pordeus, 2005; Faria and Weiner, 2005; Teixeira et al., 2008a, b in press). Furthermore, dealing specifically with it has been shown that the allergic reactions to foods may provoke characteristic responses in the skin, gastrointestinal and
respiratory tract, regardless of the immunopathogenic mechanism responsible for the reaction (Sampsom et al., 2005). As such allergic responses require complex interactions between the protein and the immune system, which are notoriously difficult to predict (Huby et al., 2000). There has been considerable recent broadening of basic concepts of intestinal food allergy, in particular the importance of non-IgE-mediated mechanisms. The traditional emphasis on IgE-mediated allergy now appears inappropriate in light of current studies of the basic mechanisms of OT to dietary antigen and of increasing recognition of the requirement for early infectious challenge in the prevention of allergic sensitization (Chehade, 2007). This major change in emphasis has been elicited both by basic scientific studies and by the recognition of novel patterns of food allergic diseases within the pediatric population. A rapid increase in food-allergic sensitization has been noted in the last decade in special, two previously rare phenomena: multiple food allergies and sensitization of exclusively breast-fed infants to antigens eaten by the mother have become commonplace (Murch, 2000). In our study, we propose that prior systemic immunization associated to the port of entry of the specific antigen is a strong cause of induction of chronic inflammatory reactions of the gut. Although other sensitizing mechanisms that simulate the clinical setting such as skin sensitization (Strid et al., 2005) or oral sensitization associated to cholera toxin (Mayer, 2005) could have been used, we chose to sensitize animals using the subcutaneous rout in order to quantify the exact amount of antigen that was given. In our results, the group of animals which received OVA before (E1) the initiation of the 4-week peanut CD presented specific Ab titers similar to the tolerant group. Animals which received free access to OVA during the first and second week of the CD (E2 and E3) presented specific Ab titers similar to the primary immunization of immune control animals. As of the third week of CD, we show that the specific Ab titers of the groups that received OVA during this period, groups E4 and E5, were similar booster level of the immune control group. After booster immunization group E1 continued to present low Ab titers in contrast to all other experimental groups (E2–E5) that received OVA to drink as of day 1 of CD exposure. These latter groups presented Ab titers equivalent or higher to immune control group confirming their immunological status. Thus, oral administration of OVA in mice sensitized to peanuts before the induction of chronic inflammation due to the exposure to a CD containing the specific antigen resulted in the development of OT. In contrast, as of the exposure to the CD which induces significant inflammatory changes to the gut, the introduction of the
ARTICLE IN PRESS P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
novel protein led to a shift from the induction of OT to systemic immunization. Although we have not determined the profile of the cells that infiltrate the gut during CD we could argue that these are cells with an inflammatory profile which have been elicited during systemic immunization and that migrate to the gut upon exposure to the specific antigen inducing a shift from a tolerogenic to an immunizing environment in the gut. In contrast to current thought that peanut allergy is primarily a Th2-driven process, there is accumulating data from the literature indicating that this food allergy may be a result of a mixed Th1Th2 immune response. The absence of systemic anaphylactic reactions, the presence of a mononuclear infiltrate with few mast cells and eosinophils in our murine model corroborates to the idea that this is a mixed Th1Th2 modulation of peanut allergy. Finally, the verification of similar inflammatory responses in the gut of mice with different genetic backgrounds (Balb/c and C57BL/B6) provides an interesting method for the study of this mixed immune response to peanut proteins. The animals that were submitted to a chronic gut inflammatory process showed more heterogeneity in the individual immune responses when compared to systemic immunization. This may be due to the fact that the animals were free to determine the amount of daily intake. The intrinsic potential of protein allergens to induce sensitization will be manifest only in susceptible individuals, and then only if the allergen is encountered in sufficient quantities and via a relevant route of exposure (Huby et al., 2000). In our model this fact can be seen in the animals that are immunized (thus susceptible) but not challenged in the diet in which the introduction of a novel protein leads to OT due to normal gut environment. Preliminary results indicate that during the recovery period of the chronic inflammation of the gut a concomitant recovery of the ability to induce OT is shown arguing that the deregulation of the normal physiology is transitory depending on local exposure to the allergen. In conclusion the induction of OT or systemic immunization to a novel protein depends on the local environment of the gut. Taking these results to the clinical setting we could suggest that children with gut alterations should not be exposed to novel proteins as a measure to avoid multiple food allergies.
References Bruijnzeel-Koomen, C., Ortolani, C., Aas, K., BindslevJensen, C., Bjo¨rkste´n, B., Moneret-Vautrin, D., Wu¨thrich, B., 1995. Adverse reactions to food. European academy of allergology and clinical immunology subcommittee. Allergy 50, 623–635.
277
Chehade, M., 2007. IgE and non-IgE-mediated food allergy: treatment in 2007. Curr. Opin. Allergy Clin. Immunol. 7, 264–268. Engelmann, P.A., Beyer, K., Wesley Burks, A., Lack, G., Liacouras, C.A., Hourihane, J.O., Sampson, H.A., Sodergren, E., 2008. New visions for food allergy: an iPAC summary and future trends. Pediatr. Allergy Immunol. (Suppl. 19), 26–39. Faria, A., Weiner, H., 2005. Oral tolerance. Immunol. Rev. 206, 232–259. Huby, R., Dearman, R., Kimber, I., 2000. Why are some proteins allergens? Toxicol. Sci. 55, 235–246. Husby, S., Mestecky, J., Moldoveanu, Z., Holland, S., Elson, C., 1994. Oral tolerance in humans. T cell but not B cell tolerance after antigen feeding. J. Immunol. 152, 4663–4670. Johansson, S.G., Hourihane, J.O., Bousquet, J., BruijnzeelKoomen, C., Dreborg, S., Haahtela, T., Kowalski, M.L., Mygind, N., Ring, J., van Cauwenberge, P., van HageHamsten, M., Wu¨thrich, B., 2001. EAACI (the European academy of allergology and cinical immunology) nomenclature task force. A revised nomenclature for allergy. An EAACI position statement from the EAACI nomenclature task force. Allergy 56, 813–824. Kaminogawa, S., Nanno, M., 2004. Modulation of immune functions by foods. Evid-Based Complement. Altern. Med. 1, 241–250. Kraus, T.A., Toy, L., Chan, L., Childs, J., Cheifetz, A., Mayer, L., 2004a. Failure to induce oral tolerance in Cronh’s and ulcerative colitis patients: possible genetic risk. Ann. NY Acad. Sci. 20, 225–238. Kraus, TA., Toy, L., Chan, L., Childs, J., Mayer, L., 2004b. Failure to induce oral tolerance to a soluble protein in patients with inflammatory bowel disease. Gastoenterology 126, 1771–1778. Kuchroo, V., Das, M., Brown, J., Ranger, A., 1995. B7-1 and B7-2 costimulatory molecules activate differentially the Th1/Th2 developmental pathways: application to autoimmune disease therapy. Cell 80, 707–718. Mayer, L., 2005. Mucosal immunity. Immunol. Rev. 206, 5–21. Mayer, L., Shao, L., 2004. Therapeutic potencial of oral tolerance. Nat. Rev. 4, 1–14. Murch, S., 2000. The immunologic basis for intestinal food allergy. Curr. Opin. Gastroenterol. 16, 552–557. Neurath, M., Finotto, S., Glimcher, L., 2002. The role of Th1/Th2 polarization in mucosal immunity. Nat. Med. 8, 567–573. Nowak-Wegrzyn, A., 2003. Future approaches to food allergy. Pediatrics 111, 1672–1680. Ortolani, C., Bruijnzeel-Koomen, C., Bengtsson, U., BindslevJensen, C., Bjo¨rkste´n, B., Høst, A., Ispano, M., Jarish, R., Madsen, C., Nekam, K., Paganelli, R., Poulsen, L.K., Wu¨thrich, B., 1999. Controversial aspects of adverse reactions to food. European academy of allergology and clinical immunology (EAACI) reactions to food subcommittee. Allergy 54, 27–45. Sampsom, J., Van Berkel, L., Van Helvoort, J., Unger, W., Jansen, W., Thepen, T., Mebius, R., Verbeek, S., Kraal, G., 2005. FcgRIIB regulates nasal and oral tolerance: a role for dendritic cells. J. Immunol. 174, 5279–5287.
ARTICLE IN PRESS 278
P.O. Paschoal et al. / Immunobiology 214 (2009) 269–278
Scott, H., Sampson, H., 2006. Food allergy. J. Allergy Clin. Immunol. 117, 470–475. Strid, J., Hourihane, J., Kimber, I., Callard, R., Strobel, S., 2005. Epicutaneous exposure to peanut protein prevents oral tolerance and enhances allergic sensitization. Clin. Exp. Allergy 35, 757–766. Teixeira, G., 2003. Um Modelo Murino de Inflamac¸a˜o Intestinal Croˆnica Toleraˆncia e Imunizac¸a˜o Oral: antigenicidade versus Temporalidade. Universidade Federal Fluminense, Nitero´i, p. 213. Teixeira, G., Antunes, D., Castro Jr., A., Campos, S., Pedruzzi, M., Paschoal, P., Costa, J., Arau´jo, V., Andrade, L., Moreno, S., Caldas, L., Cardoso, G., Nobrega, A., 2008a. Induction of an antigen specific gut
inflammatory reaction in mice and rats: a model for human inflammatory bowel disease. Braz. Arch. Biol. Technol. (in press). Teixeira, G., Paschoal, P., Oliveira, V., Pedruzzi, M., Campos, S., Andrade, L., No´brega, A., 2008b. Diet selection in immunologically manipulated mice. Immunobiology 213, 1–12. Vaz, N., Pordeus, V., 2005. Visita a` imunologia. Arq. Bras. Cardiol. 85, 350–362. Worbs, T., Bode, U., Yan, S., Hoffmann, M., Hintzen, G., Bernhardt, G., Foster, R., Pabst, O., 2006. Oral tolerance originates in the intestinal immune system and relies on antigen carriage by dendritic cells. J. Exp. Med. 203, 519–527.