Perinatal Events in Relation to Sensitization in the Human

BENGT BJÖRKSTÉN

Division of Pediatrics, Department of Health and Environment, Linköping University and Center for Allergy Research, Karolinska Institutet, Stockholm, Sweden

	INTRODUCTION

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

Pregnancy is associated with a strong skewing towards helper T cell type 2 (Th2)-type immunity and if this does not occur in time, there is an increased risk for abortion (1). As a consequence, neonatal immunity is Th2-skewed and allergen- specific T cell responses are common already at birth (2). The unanswered question is still how the fetus is exposed to the inhalant allergens in utero, nor do we know to what extent these responses are normal and protective or indicate sensitization and future allergy. The difference between individuals with and without a genetically determined atopic propensity may be how readily this immunity deviates toward a Th1-type response after birth (3). The atopic phenotype is associated with a prolonged period of Th2-type immune responses to allergens early in life (2). These findings confirm on the T cell level previous observations that temporary low-level IgE responses to food and inhalant allergens are common during the first years of life, but are then downregulated in nonatopic individuals and continue to increase in children who develop allergic manifestations (4). Thus, both atopic and nonatopic individuals have immune responses to allergens early in life, although the type of immunity differs. Therefore, it seems reasonable to look for postnatal environmental factors that downregulate Th1-type immune responses and thus prevent the development of allergy, rather than to advocate allergen avoidance during pregnancy. A search for factors that enhance the induction of tolerance would seem to be better than to devote all interest toward risk factors enhancing sensitization.

	OBSTETRIC ASPECTS

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

There are several reports that premature birth is associated with a reduced risk for allergic disease. For example, in a large analysis of perinatal risk factors for allergy in 148,000 Swedish conscripts for military service, the odds ratio for hay fever was 0.85 (p < 0.001) for men who were born after 33-36 gestational weeks and 0.75 for men born even more prematurely (5). Maternal age at birth also affected the likelihood for hay fever, as young mothers were less likely than older mothers to have allergic children. These findings are of interest in the light of a report that allergic women may have more children as compared with nonallergic women (6). As a normal pregnancy is associated with a Th2-skewed immunity (1), it is possible that this occurs more readily in atopic women. Consequently, a propensity toward Th2-type immunity would be advantageous in connection with pregnancy.

	THE ROLE OF BREAST MILK

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

During early infancy there is a close immunological interaction between the mother and her offspring through the breast milk, but relatively little is known about the exchange of immunological information. Human milk not only provides passive protection against infections, but also actively stimulates infant immunity, for example, IgA antibody synthesis and cell-mediated immunity, and it contains several cytokines (7). There are considerable individual variations in the composition of human milk, however. This may explain the controversy with regard to the possible allergy-preventive effects of breast-feeding. If individual variations in the milk modulate the development of immunity in the neonate, then maternal immunity may represent an environmental factor, which would influence the risk for allergic manifestations in her child, possibly even several years later. Similar to the antiinfective properties, an antiallergic effect by secretory IgA (sIgA) antibodies in human milk has been proposed. Low levels of total IgA and IgA antibodies against cow's milk have been reported in milk from mothers of food-allergic infants (8), while in a larger, prospective study this was not confirmed (9).

It has been suggested for many years that at least in some patients atopic disease is associated with an altered fatty metabolism (10).

Human milk is the major source of essential fatty acids and it contains high levels of long-chain fatty acids (11). Furthermore, the newborn baby lacks the capacity to elongate its long-chain metabolites (12). A disturbed composition of polyunsaturated fatty acids (PUFAs) has been reported in milk from mothers of atopic infants. For example, the levels of the essential fatty acid linoleic acid (LA) as well as its metabolites are lower in milk from atopic, as compared with nonatopic, mothers and the ratio between n-6 and n-3 fatty acids is higher (13).

Differences in the fatty acid composition of human milk seem to have relevance for breast-fed infants. Thus, lower levels of the n-3 fatty acids eicosapentaenoic acid (EPA), docosapentaenoic acid (DPA), and docosahexaenoic acid (DHA) have been observed in mothers of infants who developed allergic disease during the first year of life as compared with mothers of babies who did not develop any allergic manifestations (10). These differences were independent of maternal allergy. Similarly, low levels of certain polyamines have been observed in milk from atopic mothers (10). As these compounds are required for optimal DNA synthesis, variations in the levels could affect immune responses to foreign antigens in breast-fed infants.

	MICROBIAL EXPOSURE

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

It is well established that infections and other microbial pressure may drive the immune system toward a Th1-like response. It has been suggested that recurrent infections would enhance skewing toward Th1 type responses by exerting a pressure on the immune system (3). These suggestions are based on indirect evidence, for example, an inverse relationship between allergy prevalence and the number of siblings and congested living conditions. The evidence of a relationship between respiratory tract infections and protection against allergy is circumstantial, however, and not supported by comparisons of risk factors and allergy prevalence in different regions.

The intestinal microflora is the most likely source for microbial pressure that would enhance Th1-type responses. It has been shown in animal experiments that germ-free pups have a prolonged period of Th2 immune responses, including IgE antibody formation and a delayed development of oral tolerance (14). When the animals are colonized with bacteria of the normal commensal intestinal flora, oral tolerance and immune deviation toward Th1 responses rapidly develop. Studies have shown that the intestinal microflora differ in Estonian and Swedish infants, that is, in two countries with a low and a high prevalence of atopic disease, for example, more lactobacilli and less clostridia in the former (15). Differences in the microbial flora of infants in the two countries are present already during the first week of life. Thus, the postnatal colonization seems to be more intense in Estonian than in Swedish babies, as the counts of aerobes, particularly staphylococci, enterococci, and enterobacteria, are much higher in the former (16). At 1 mo the bacterial counts in the Swedish babies have reached levels similar to those in the Estonian infants. At that age, however 80% of Estonian infants, as compared with only 30% of Swedish babies, are colonized with lactobacilli.

There are also observations showing differences in the composition of the microflora between allergic and nonallergic children (17). These findings are supported by a Swedish comparative analysis of short-chain fatty acids in stool samples from 25 allergic and 47 nonallergic infants (18). Isocaproic acid was detected almost exclusively in allergic infants. This compound is associated with the presence of Clostridium difficile. In contrast, the levels of several fatty acids associated with a Lactobacillus flora were higher in the nonatopic infants. It is tempting to suggest that differences in the indigenous intestinal flora might effect the development and priming of the immune system in early childhood and that the observed differences between Estonian and Swedish infants and between allergic and nonallergic children are not secondary phenomena.

	ALLERGY PREVENTION IN EARLY INFANCY

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

Previous efforts to prevent allergy by various avoidance regimens applied from birth have at best had only marginal effects. This may be due to the complexity and demands of the intervention, resulting in a low level of compliance, or simply because allergen exposure and environmental pollution are not major risk factors to be addressed in primary prevention, although they may be so in secondary prevention. Future strategies should therefore be based on new concepts based on current knowledge of how tolerance and sensitization develop. A strategy could be envisaged by which immune deviation toward a Th1 type of immune response to inhalant allergens could be induced, for example, by administering allergen and suitable adjuvant in a vaccination program begun early in life. A more general possible prevention strategy could include modification of the microbial environment, particularly the intestinal microflora, which is probably the major source of microbial stimulation of the immune system in early infancy. Such studies, using well-defined strains of Lactobacillus, are currently ongoing.

A word of caution is appropriate, however. While allergists would favor skewing of Th2-type immune responses to allergens toward the Th1, researchers in autoimmune diseases would favor stimulation of Th2-type responses as Th1 immunity is associated with disease. In Estonia, for example, not only is the prevalence of childhood allergy low, so is the prevalence of type 1 diabetes (19) and celiac disease (20). These observations may indicate that merely stimulating one type of immunity may not be optimal, but rather a balanced Th1/Th2 immunity should be the goal.

	CONCLUSION

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

The major lifestyle and environmental factors triggering sensitization probably exert much of their influence during the first few years of life or even in early infancy. The conclusion from a broad compilation of what is known regarding triggers of allergy is that we cannot identify which environmental factors are the major reasons for the large increase in the prevalence of allergy in recent years. The mother not only provides genetic factors to her offspring but she is also an important environmental factor during the gestational and lactation periods. Individual variations in maternal immunity and the composition of breast milk may therefore affect the likelihood of allergy in her offspring. The relationship between allergy development and the development of the intestinal microflora, which starts immediately after birth, deserves a closer analysis. Future research should be truly interdisciplinary and "lifestyle" should be given a broad interpretation.

	DISCUSSION

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

Neijens: You found higher counts for staphylococci in stools. These are, of course, very potent immunostimulators by superantigen mechanisms. Is the number higher also on the skin?

Björkstén: We are indeed interested in these staphylococci, as we believe that lipopolysaccharide is a key driving force. We have no data from the skin, however. But I want to be very clear on this: I am not saying that lactobacilli are good or bad, or that staphylococci are bad. I am presenting a phenomenon and I am saying there is a major ecological difference. We are now working in germ-free animals, but we still do not know which of the bacteria are good and which are bad. I also have to add that 90% of the intestinal flora is unknown.

Platts-Mills: What happens to cord blood T cells in children born by cesarian section? We heard about the remarkable rates of cesarian sections in Perth. What happens with children born by cesarian section with fecal colonization?

Björkstén: We don't know. We don't have a high prevalence of cesarian sections, however.

Platts-Mills: The other issue is with the cat dose response. You made the suggestion that having a cat at home decreases your risk of atopy. Do you think that is a new phenomena, is there any evidence that this has changed?

Björkstén: There clearly is a problem in view of the political implications. We have seen it before, but we never made a story out of it.

Platts-Mills: I am not doubting the data at all, it is also clear from our study in America that the dose of cat in the house is not a predictor of sensitization of the children. It is also clear in our study in Kiruna that the presence of a cat in the house is a negative risk factor for asthma. The question is: Is this really a mechanism of excessive antigen dose? Do you think the situation in Sweden has changed?

Björkstén: I don't know.

Quanjer: Your study seems to suggest that greater exposure to cat allergen provides protection against development of atopy. Dr. Brunekreef made a similar observation, but he thought it might be attributed to selection, a sort of healthy worker effect, in the sense that families with a history of atopy did not have cats. Maybe Dr. Brunekreef would care to comment?

Brunekreef: The study you are referring to is a questionnaire-based study, with all its obvious limitations. What we found was indeed an inverse relationship between reported pet allergy and associated symptoms and pet ownership. We asked a number of questions about the years preceding the questionnaire study, which was among 6- to 12-yr-old children. We found that if you take pet-avoidance behavior into account, the relationship actually reversed into a positive relation between pet ownership and reported respiratory symptoms and reported pet allergy. We concluded from that study that there was at least a strong indication that parents were actually reporting pet avoidance and that the reported pet allergy and symptoms were also higher in the groups that reported never having had pets at all. So, they were avoiding it from the very beginning for reasons of problems in the parents or other children. We asked all the families without pets why they did not have pets. If they brought up anything related to allergy we excluded them from the analysis. The asthma came down from 10 to 8.8% and the skin text positivity from 14.1 to 11.2%. But the effect was still significant. So, we tried our best to avoid this pitfall. We put a different interpretation on it, which is that the distribution of cat allergen within the community is so widespread and the transport of cat allergens into houses without a cat is quite sufficient to sensitize genetically programmed children. In northern Sweden we think they are all exposed to cat regardless of whether they have a cat in the house.

Björkstén: Absolutely. We still have to find a single house in Linköping without detectable cat allergen. The levels are of course higher when there is a cat at home.

Von Mutius: For cats this is indeed an issue. Contact with other animals could be important. Studies are known in which the exposure to, for example, pig has a strong protective effect. Furthermore, children grown up on farms have a significantly lower prevalence of atopy.

Björkstén: There are, of course, alternative explanations. For example, when you believe in the effect of the intestinal flora you can say that the ecology would be different when exposed to indoor animals. In Estonia we only looked at urban people, we don't have data on a rural population. You can look at children from atopic families who are not atopic themselves, despite being exposed to high levels of cat in order to study the immune response in children who might have been toleralized in some way. This is part of an ongoing prospective study.

	Footnotes

Correspondence and requests for reprints should be addressed to B. Björkstén, M.D., Center for Allergy Research, Karolinska Institute, Nobels väg 5, 171 77 Stockholm, Sweden. E-mail: bengt.bjorksten{at}admin.ki.se

	References

TOP INTRODUCTION OBSTETRIC ASPECTS THE ROLE OF BREAST... MICROBIAL EXPOSURE ALLERGY PREVENTION IN EARLY... CONCLUSION DISCUSSION REFERENCES

1. Wegmann, T., H. Lin, L. Guilbert, and T. Mosmann. 1993. Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a Th2 phenomenon? Immunol. Today 14: 353-356 [Medline].

2. Prescott, S., C. Macaubas, T. Smallacombe, B. Holt, P. Sly, and P. G. Holt. 1999. Development of allergen-specific Th2 responses persist in atopic and normal children. Lancet 353: 196-200 [Medline].

3. Holt, P., P. Sly, and B. Björkstén. 1997. Atopic versus infectious disease in childhood: a question of balance? Pediatr. Allergy Immunol. 8: 1-5 [Medline].

4. Hattevig, G., B. Kjellman, and B. Björkstén. 1993. Appearance of IgE antibodies to ingested and inhaled allergens during first 12 years of life in atopic and non-atopic children. Pediatr. Allergy Immunol. 4: 182-189 [Medline].

5. Bråbäck, L., and A. Hedberg. 1998. Perinatal risk factors for atopic disease in conscripts. Clin. Exp. Allergy 28: 936-942 [Medline].

6. Nilsson, L., N.-I. Kjellman, and B. Björkstén. 1997. Parity among atopic and non-atopic mothers. Pediatr. Allergy Immunol. 8: 134-137 [Medline].

7. Goldman, A.. 1986. Immunologic system in human milk. J. Pediatr. Gastroenterol. Nutr. 5: 343-345 [Medline].

8. Machtinger, S., and R. Moss. 1986. Cow's milk allergy in breast-fed infants: the role of allergen and maternal secretory IgA antibody. J. Allergy Clin. Immunol. 77: 341-347 [Medline].

9. Fälth-Magnusson, K., and N.-I. M. Kjellman. 1992. Allergy prevention by maternal elimination diet during late pregnancya 5-year follow-up of a randomized study. J. Allergy Clin. Immunol. 89: 709-713 [Medline].

10. Duchén, K. 1999. Human milk factors and atopy in early childhood. Linköping, Sweden.

11. Decsi, T., and B. Koletzko. 1994. Polyunsaturated fatty acids in infant nutrition. Acta Paediatr. Suppl. 395: 31-37 .

12. Luukkainen, P., M. Salo, J. Visakorpi, N. Räihä, and T. Nikkri. 1996. Impact of solid food on plasma arachidonic and docosahexaenoic acid status of term infants at 8 months of age. J. Pediatr. Gastroenterol. Nutr. 23: 229-234 [Medline].

13. Duchén, K., G. Yu, and B. Björkstén. 1998. Atopic sensitization during the first year of life in relation to long chain polyunsaturated fatty acid levels in human milk. Pediatr. Res. 44: 478-484 [Medline].

14. Sudo, N., S. Sawamura, K. Tanaka, Y. Aiba, C. Kubo, and Y. Koga. 1997. The requirement of intestinal bacterial flora for the development of an IgE production system fully susceptible to oral tolerance induction. J. Immunol. 159: 1739-1745 [Abstract].

15. Sepp, E., K. Julge, N. Vasar, P. Naaber, B. Björkstén, and M. Mikelsaar. 1997. Intestinal microflora of Estonian and Swedish infants. Acta Paediatr. 86: 956-961 [Medline].

16. Sepp, E., P. Naaber, T. Voor, M. Mikelsaar, and B. Björkstén. 2000. Development of intestinal microflora during the first months of life in Estonian and Swedish infants. J. Microbiol. Ecol. Health 12: 22-26 .

17. Björkstén, B., P. Naaber, E. Sepp, and M. Mikelsaar. 1999. The intestinal microflora in allergic Estonian and Swedish 2-year old children. Clin. Exp. Allergy 29: 342-346 [Medline].

18. Böttcher, M., A. Sandin, E. Norin, T. Mitvedt, and B. Björkstén. 2000. Microflora associated characteristics in faeces from allergic and non-allergic children. Clin. Exp. Allergy (In press)

19. Tuomilehto, J., T. Podar, A. Reunanen, I. Kalits, R. Lounamaa, and E. Tuomilehto-Wolf. 1991. Comparison of incidence of IDDM in childhood between Estonia and Finland, 1980-1988. Diabetes Care 14: 982-988 [Abstract].

20. Uibo, O.. 1994. Efficacy of the IgA-class antigliadin antibody test in search for new cases. J. Pediatr. Gastroenterol Nutr. 18: 53-55 [Medline].