Infections prevent the development of asthma - NCBI

JOURNAL OF THE ROYAL SOCIETY OF MEDICINE October 1999

Volume 92

Number 10

ISSN 0141-0768

Infections prevent the development of asthmatrue, false or both? In the complex aetiology of asthma, genetic and environmental mechanisms operate in concert. The remarkable increases in prevalence of allergic respiratory diseases over the past 20-30 yearsl 2, and differences between communities with similar genetic make-up3'4, indicate the importance of environmental factors. Ten years ago Strachans proposed that the increase in prevalence of atopic disease in westernized communities might be due to improved hygiene and hence fewer childhood infections. There is, however, the opposing argument that some childhood viral infections are strongly associated with the later development of recurrent wheeze and atopy in childhood6; thus, viral infection in early childhood may promote the development of recurrent wheezing in later life. In this editorial we suggest mechanisms whereby infections can both promote and prevent the establishment of asthma under different circumstances. Although changes in exposure to respiratory irritants, allergens or diet are undoubtedly important causes of the increased prevalence, here we focus on the role of infection. A CAUSAL ROLE FOR INFECTIONS IN THE PATHOGENESIS OF ASTHMA

Respiratory viruses are now considered the most important triggers for acute exacerbation of established asthma7. Exacerbation is thought to be caused by altered cytokine responses in asthmatics, reflected by prolonged eosinophil accumulation in the lungs following viral infection8 and enhanced penetration of inhaled allergens during virusinduced lung inflammation9. Both clinical and laboratory data suggest an additional causal link between bronchiolitis and asthma. Sigurs6 reported 47 infants admitted to hospital with respiratory syncytial virus (RSV) bronchiolitis at a mean age of 3.5 months. Of these, 23% were diagnosed with asthma in the 3 ensuing years, compared with 1% of controls. A positive test for serum IgE antibodies was found in 32% of the bronchiolitics compared with 9% of controls. An association between infantile bronchiolitis and the diagnosis of asthma was also found by Sly10 in 92% of 48 patients

followed up prospectively for 5 years. Murray1I reported wheezing in 43% of children previously admitted to hospital with infantile bronchiolitis compared with 15% of controls, in a study of 73 children followed up for 5.5 years. Respiratory symptoms may persist for 8-13 years in children after RSV bronchiolitis12, and histamine responsiveness demonstrates hyperreactivity 10 years or more after the event13. Although these studies may point to a causal link between RSV bronchiolitis and asthma, the possibility of a genetic predisposition to both asthma and bronchiolitis can not be ruled out. Work in animals supports a causal link between viral bronchiolitis and asthma. Brown Norway rats develop chronic, episodic and reversible airway obstruction after bronchiolitisI4. A predominant CD4+ T cell response with reduced interferon-gamma (IFN-y) production is seen, leading to airway remodelling with persistent inflammation, fibrosis and deposition of extracellular matrix material. A mouse model of RSV bronchiolitis demonstrates a T helper 2 (Th2) cell predominance in response to the major surface glycoprotein G of RSV in certain mouse strains15. This is in contrast to the usual T helper 1 (Thl) response seen in most primary viral infections, leading to prominent lung eosinophilia after RSV challenge. This response is linked to the inability of the G-protein to stimulate CD8+ T cells, emphasizing the important non-cytolytic role of CD8+ cells. CD8+ T cells also exhibit type 1 and type 2 cytokine production patterns (Tcl and Tc2, respectively) and may have an important role in determining CD4 phenotypes16. When individuals are exposed to respiratory viruses and allergens simultaneously, interactions have been reported that could lead to the establishment of asthma. The combination of atopy and concurrent viral illness is a greater risk factor than any one of these factors alone in children presenting to an emergency room with acute airway obstruction17. Experimental rhinovirus-16 inoculation in individuals with allergic rhinitis alters their response to allergen bronchoprovocation to favour the development of the late-phase response18. This is associated with airway inflammation and hyperresponsiveness and suggests that viruses may induce the development of an asthmatic phenotype under specific conditions. Work in animals supports this concept. Uninfected mice develop no IgE antibody against inhaled antigen (ovalbumin), but during acute influenza infection sensitization does take place19. Neither influenza A virus nor ovalbumin alone causes changes in serum IgE or airway responsiveness to inhaled

0 [-'

495

JOURNAL OF THE ROYAL SOCIETY OF MEDICINE

metacholine but both are increased when antigen exposure coincides with influenza infection20. O'Donnell observed acute systemic illness and collapse in response to cutaneous ovalbumin administration in mice previously exposed to concurrent RSV infection and nebulized ovalbumin, but not in animals exposed to either of these insults alone9. This suggests that viral inflammation allows inhaled antigen to penetrate the barrier of the respiratory mucosa, promoting sensitization. Chronic Chlamydia pneumoniae infection has been linked to acute exacerbations of asthma21 but immune responses of patients with frequent exacerbations additionally suggest a role for Chlamydia in asthma pathogenesis. Cunningham investigated 108 children with asthma and found a tendency to remain polymerase-chain-reaction-positive for Chiamydia in those with frequent acute episodes, suggesting chronic infection. Previous chlamydial exposure has been demonstrated by Cook et al. in 34.8% of patients with severe chronic asthma22 but other studies23 provided no evidence of chlamydial disease in asthmatics. Evidence of Mycoplasma pneumoniae infection has been found in the lower airways of stable asthmatics with greater frequency than in control subjects, suggesting a pathogenic role24. These issues require further investigation since most studies are small and there are many conflicting findings. As with the RSV data, a simultaneous genetic predisposition to both asthma and these infections cannot be ruled out. The mechanisms by which viral and other infections during early life could affect later development of lung disease were summarized by Openshaw and Lemanske25. These include: (1) virus chronicity, persistence or latency; (2) permanent changes to local anatomy; (3) impaired lung growth; (4) local foothold for other acute or chronic infections; (5) immunological tolerance; and (6) establishment of local immunological 'mood'. Sensitization to aeroallergens during respiratory infections can be added to this list. THE POSSIBLE PROTECTIVE ROLE OF INFECTIONS IN THE ESTABLISHMENT OF ASTHMA A protective role of early life infections in the development of atopy and asthma was suggested by British cohort studies conducted since 1958. A consistent relationship was found

496

between birth order and the risk of atopy, the presence of older siblings appearing to protect younger children26'27'28. In 1989, Strachan29 reported that repeated infections in early life might prevent hayfever. The increase in atopic disease in developed countries was ascribed to the decreased cross-infection rates due to decreased family size29, now often referred to as the hygiene effect. Bodner et al. reported data obtained from a 1964 cross-sectional survey of Aberdeen schoolchildren. Although they too found the

Volume 92

October 1 999

protective effect of large sibships,

a higher number of infections before the age of three years was associated with a significant trend in odds ratios towards an increased risk of asthma. Remarkably, measles after the age of three protected against asthma, but chickenpox increased the risk and rubella was associated with eczema30. Working in Tucson, Arizona, Martinez examined the relations of lower respiratory tract illnesses (LRIs) in the first three years of life to atopy and serum IgE levels. Non-wheezing LRIs were associated with lower IgE levels and lower skin test reactivity than wheezing LRIs and absent LRIs. At the age of nine months, children with non-wheezing LRIs had higher IFN-y production from mononuclear cells than the other groups. Martinez argues that the immune response to viral infections in these children leads to selection of Thl cells with suppression of Th2 cells31. Von Mutius et al. conducted surveys soon after the reunification of East and West Germany, and showed a higher frequency of respiratory symptoms in the East German children (42% versus 27%), but a lower prevalence of asthma (3.8% versus 5.4%)32. These findings have been used to support the argument that respiratory infections protect against asthma, but there are several alternative interpretations. A high percentage of children in East Germany were diagnosed as having wheezy bronchitis, perhaps reflecting different definitions in the two regions. Furthermore, the same skin test antigens may not be appropriate in the two regions. Others have not been able to confirm the sibship effect. Farooqi and Hopkins33 found maternal atopy, antibiotic use in the first two years of life and administration of whole-cell pertussis vaccine to be predictors of atopy in their retrospective study of 1934 individuals from a 1974-1984 birth group at a family doctor practice. They found no association for sibship size. Another interesting observation relates to the high prevalence of asthma not only in developed communities but also in African cities4 and amongst low-income minority groups in the USA34. It is unlikely that poor urban children in Africa are subjected to less crowding and therefore experience less infection than their rural counterparts. Another important study was recently reported by Shirakawa and co-workers in Japan35. In a BCG-vaccinated population, strong tuberculin responders were one-half to one-third as likely as non-responders to have asthma symptoms, and remission of asthma symptoms during childhood was six to nine times more likely. Serum IgE and Th2 cytokine levels were lower, but Thl cytokines higher, in strong responders. Shirakawa et al. argue that this reflects a causal link and not a fixed determination of both atopy and tuberculin response by genetic factors since the tuberculin response changed from positive to negative in some individuals and vice versa. There has been a marked decline in

positive tuberculin

coinciding

responders over 20 years in

with a decline in clinical tuberculosis

the

area,

cases 35

.


A study from Guinea-Bissau showed that children who had previously had measles were less likely to develop atopy than children who had been vaccinated against measles36. The protective effect of measles was also supported by the Aberdeen study30 but here there was no effect or even a positive correlation with other childhood infections and atopy. This is surprising since crowding and sibship size should have the same effect on the transmission of these infections as on measles. The other infection that has been linked to a protective effect is hepatitis A37, which is transmitted via the faecal-oral route; this would support the hygiene hypothesis. What are the proposed immunological mechanisms for the protective effect of infections in the development of atopic disease? Atopy is related to the expression of allergen-specific responses with production of Th2 cytokines such as IL-4 and IL-5, which promote IgE production and eosinophilia38. In non-atopic individuals, the T cell system is biased to a Thl phenotype with production of IFN-y and inhibition of Th2 cells. Further knowledge of the development of allergen-specific T cell memory is important for the possible primary prevention of allergic

Volume 92

October 1 999

disease. The existence of fetally-derived allergen-reactive T cells exhibiting a Th2 phenotype indicates intrauterine T cell priming39. Prescott et al. have demonstrated the continuation of these fetal allergen-specific Th2 responses during infancy with a decreased capacity to produce IFN-y in atopic neonates40. They argue that environmental factors such as microbial agents exert their effects during early life and aid in the maturation of the adaptive immune response into a Thl-dependent system. Romagnani41 suggests that the rising prevalence of atopy is related to a reduction in childhood infections such as tuberculosis, with a resultant reduced production of cytokines antagonistic to Th2 cell differentiation. Animal studies have supported this concept of immunomodulation by microorganisms. Erb et al.42 were able to suppress allergen-induced airway eosinophilia in mice by previous BCG infection a phenomenon that depended on the ability of the animals to produce IFN-y. Wang and Rook inhibited the established allergic response to ovalbumin in mice with killed Mycobacterium vaccae43. The absence of microbacterial flora from the gut inhibits the maturation of the Thl immune mechanisms and leads to lifelong bias towards Th2 responses44, an observation that

Infant with genetic predisposition to atopy Tcell phenotype biased towards type 2 cytokine production 'PROTECTIVE' infections 4*Acquired from older sibling *Mycobacteria (soil, air, water, BCG vaccine) Lack of IL-12 productionb -Measles APC'S | IL-12 prduction by APCs and Lack of IFN-y production 'Hepatitis A dendritic cells. 'GI flora IL 4 and IL 5 production via IFN-y production by CD4+, peristence of immature immune 4

4

I

CD8+ T cells and NK cells

TD-biased T1cell

Damaging infections (even in absence } of genetic predisposition) ViRalC pneRuongaerg pneumoniaexposu -V;ira RTI during allergen exposure J

LACK of 'protective' infections *Few siblings *Excessive 'hygiene' *Antibiotics in first two years of life -Vaccination and prevention of some diseases (e.g. measles)

system

I

type

Established Th2 cytokine environment A

r 'Viral chronicity *Impaired lung growth 'Epithelial damage and other permanent anatomical changes 'Immunological tolerance or 'mood'

Figure 1 Interactions between infections and development of atopy. Gl=gastrointestinal; RTI=respiratory tract infection; APCs=antigenpresenting cells

497


offers interesting support to the data on antibiotic use during the first two years of life as a risk factor for the development of atopy33. The immunological basis of the hygiene hypothesis can be summarized as follows (Figure 1). Genetically predisposed individuals are born with allergen-specific T cells exhibiting a Th2 phenotype. Immunomodulatory microorganisms such as intracellular bacteria or viruses induce macrophages and dendritic cells to produce IL- 12, which induces natural killer cells and Thl cells to produce IFN-y. IFN-y provides the environment for the differentiation of antigen-specific CD4+ T cells into Thl cells and CD8+ T cells into Tcl cells with even higher IFN-y production. In the absence of such early stimuli, the Th2-biased immature immune system persists and the allergen-specific responses become established to express an atopic phenotype. CONCLUSION There is good evidence that interactions with some

respiratory viruses, particularly RSV, are followed by an increase in pulmonary symptoms. This association remains for at least 10 years and childhood acute respiratory illness may even be associated with respiratory morbidity and mortality in adult life45. Although respiratory infections have not conclusively been causally linked to later atopic disease, strong evidence from animal studies makes this highly plausible. There is equally good evidence that exposure to an antigenically rich environment helps the immune system to mature from a Th2 cytokine profile at birth to a Thi cytokine profile during later childhood. This maturation may be slow and inefficient in children predisposed to atopy40 and may be promoted by repeated exposure to ingested and inhaled bacteria and by some childhood viral infections. Changes in these parameters may, in part, account for the recent rise in atopic diseases and asthma in the developed world. The effects of infectious agents on the development of allergy and atopy are not simple: infections may be protective in some individuals and yet promote long-term symptoms in others. To adopt polar views in the debate about the benefit and harm caused by childhood infections is inappropriate, since there are evident truths on both sides.

498

Volume 92

REFERENCES 1 Burney PGJ, Chinn S, Rhona RJ. Has the prevalence of asthma increased in children? Evidence from the national study of health and growth. BMJ 1990;300:1306-10 2 Peat JK, van den Bergh RH, Green WF, et al. Changing prevalence of asthma in Australian children. BMJ 1994;308:1591-6 3 von Mutius E, Martinez FD, Fritsch C, et al. Prevalence of asthma and atopy in two areas of West and East Germany. AmJ Respir Crit Care Med 1994;149:358-64 4 Ng'ang'a LW, Odhiambo JA, Mungai MW, et al. Prevalence of exercise induced bronchospasm in Kenyan school children: an urbanrural comparison. Thorax 1998;53:919-6 5 Strachan DP. Hay fever, hygiene and household size. BMJ 1989;299:

6

7 8

9

10 11 12

13

14

15

16

17 18 19

Acknowledgments This work was supported by programme grant 054797 from the Wellcome Trust. G.W. is supported by a research fellowship from the South African Pulmonology Society, Allen and Hanbury, South Africa, and the South African MRC. Peter J M Openshaw

20

Gerhard Walzl

23

Department of Respiratory Medicine, National Heart and Lung Institute, Imperial College School of Medicine at St Mary's, London W2 1 PG, UK

Octo be r 1 999

21

22

1259-60 Sigurs N, Bjarnason R, Sigurbergsson F, et al. Asthma and immunoglobulin E antibodies after respiratory syncytial virus bronchiolitis: a prospective cohort study with matched controls. Pediatrics 1995;95:500-5 Bardin PG, Johnston SL, Pattermore PK. Viruses as precipitants of asthma symptoms: II. Physiology and mechanisms. Clin Exp Allergy 1992; 22:809-22 Teran LM, Johnston SL, Schroder JM, et al. Role of nasal interleukin-8 in neutrophil recruitment and activation in children with virus-induced asthma. Am J Respir Crit Care Med 1997;155: 1362-6 O'Donnell DR, Openshaw PJM. Anaphylactic sensitization to aeroantigen during respiratory virus infection. Clin Exp Allergy 1998;28:1501-8 Sly PD, Hibbert ME. Childhood asthma following hospitalization with acute viral bronchiolitis in infancy. Pediatr Pulmonol 1989;7:153-8 Murray M, Webb MSC, O'Callaghan C, et al. Respiratory status and allergy after bronchiolitis. Arch Dis Child 1992;67:482-7 Connochie KM, Roghmann KJ. Wheezing at 8 and 13 years: changing importance of bronchiolitis and passive smoking. Pediatr Pulmonol 1989;6:138-46 Pullan CR, Hey EN. Wheezing, asthma and pulmonary dysfunction 10 years after infection with respiratory syncytial virus in infancy. BMJ 1982;284: 1665-9 Kumar A, Sorkness R, Kaplan MR, et al. Chronic, episodic, reversible airway obstruction after viral bronchiolitis in rats. Am J Respir Crit Care Med 1997;155: 130-4 Openshaw PJM. Immunopathological mechanisms in respiratory syncytial virus disease. Springer Semin Immunopathol 1995;17:187-201 Mosmann TR, Sad S. The expanding universe of T-cell subsets: Thl, Th2 and more. Immunol Today 1996;17:73-81 Duff AL, Pomeranz ES, Gelber LE, et al. Risk factors for acute wheezing in infants and children: viruses, passive smoke and IgE antibodies to inhalant allergens. Pediatrics 1993;92:535-40 Calhoun WJ, Swenson CA, Dick EC, et al. Experimental rhinovirus 16 infection potentiates histamine release after antigen bronchoprovocation in allergic subjects. Am Rev Respir Dis 1991;144:1267-73 Sakamoto M, Ida S, Takishima T. Effect of influenza virus infection on allergic sensitization to aerosolized ovalbumin in mice. J Immunol 1984; 132:1614-17 Suzuki S, Suzuki Y, Yamamoto N, et al. Influenza A virus increases IgE production and airway responsiveness in aerosolized antigen-exposed mice. J Allergy Clin Immunol 1998;102:732-40 Cunningham AF, Johnston SL, Julious SA, et al. Chronic Chlamydia pneumoniae infection and asthma exacerbations in children. Eur RespirJ 1998;11:345-9 Cook PJ, Davies P, Tunnicliffe W, et al. Chlamydia pneumoniae and asthma. Thorax 1998;53:254-9 Larsen FO, Norn 5, Mordhorst CH, et al. Chlamydia pneumoniae and possible relationship to asthma. Serum immunoglobulins and histamine release in patients and controls. APMIS 1998;106:928-34


24 Kraft M, Cassell GH, Henson JE, et al. Detection of Mycoplasma pneumoniae in the airways of adults with chronic asthma. AmJ Respir Crit Care Med 1998;158:998-1001 25 Openshaw PJ, Lemanske RF. Respiratory viruses and asthma: can the effects be prevented? Eur RespirJ 1998;12:Suppl.27:35s-39s 26 Butland BK, Strachan DP, Lewis S, et al. Investigation into the increase in hay fever and eczema at age 16 observed between the 1958 and 1970 British birth cohorts. BMJ 1997;315:717-2 27 Strachan DP, Griffiths JM, Anderson HR, et al. Allergic sensitization and position in the sibship: a national study of young British adults. Thorax 1994;49: 1053P 28 Strachan DP, Taylor EM, Carpenter RG. Family structure, neonatal infection, and hay fever in adolescence. Arch Dis Child 1996;74:422-6 29 Strachan DP. Is allergic disease programmed in early life? Clin Exp Allergy 1994;24:603-5 30 Bodner C, Godden D, Seaton A. Family size, childhood infection and atopic diseases. Thorax 1998;53:28-32 31 Martinez FD, Stern DA, Wright AL, et al. Association of nonwheezing lower respiratory tract illnesses in early life with persistently diminished serum IgE levels. Thorax 1995;50:1067-72 32 Von Mutius E, Martinez FD, Fritsch C, et al. Prevalence of asthma and atopy in two areas of West and East Germany. AmJ Respir Crit Care Med 1994;149:350-64 33 Farooqi IS, Hopkin JM. Early childhood infection and atopic disorder. Thorax 1998;53:927-32 34 Schenker MB, Gold EB, Lopez RL, Beaumont JJ. Asthma mortality in California, 1960-1989. Am Rev Respir Dis 1993;147:1454-60 35 Shirakawa T, Enomoto T, Shimazu S, Hopkin J. The inverse relationship between tuberculin responses and atopic disorder. Science 1997; 275:77-9

Volume 92

October 1 999

36 Shaheen SO, Aaby P, Hall AJ, et al. Measles and atopy in GuineaBissau. Lancet 1996;347:1792-6 37 Matricardi PM, Rosmini F, Ferrigno L, et al. Cross sectional retrospective study of prevalence of atopy among Italian military students with antibodies against hepatitis A virus. BMJ 1997;314: 999-1003 38 Romagnani S. Induction of Thl and Th2 responses: a key role for the "natural" immune response? Immunol Today 1992;13:379-81 39 Prescott SL, Macaubas C, Holt BJ, et al. Transplacental priming of the human immune system to environmental allergens: universal skewing of the initial T-cell responses towards the Th2-cytokine profile. J Immunol 1998;160:4730-7 40 Prescott SL, Macaubas C, Smallcombe T, et a]. Development of allergen-specific T-cell memory in atopic and normal children. Lancet 1999;353: 196-200 41 Romagnani S. Regulation of the development of type 2 helper cells in allergy. Curr Opin Immunol 1994;6:838-46 42 Erb KJ, Holloway JW, Sobeck A, et a]. Infections of mice with Mycobacterium bovis Bacillus Calmette-Guerin (BCG) suppress allergeninduced airway eosinophilia. J Exp Med 1998;187:561-9 43 Wang CC, Rook GAW. Inhibition of an established allergic response to ovalbumin in BALB/c mice by killed Mycobacterium vaccae. Immunology 1998;93:307-1 3 44 Sudo N, Sawamura S-A, Tanaka K, et a]. The requirement of intestinal bacterial flora for the development of an IgE production system fully susceptible to oral tolerance induction. J Immunol 1997;159: 1739-45 45 Kiernan KE, Colley JRC, Douglas JWB, Reid DD. Chronic cough in young adults in relation to smoking habits, childhood environment and chest illness. Respiration 1976;33:36-44

499