Correspondence
Childhood Infections and Autoimmune Diseases
N Engl J Med 2002; 346:1749-1750May 30, 2002
- Article
To the Editor:
The article by Zinkernagel (Nov. 1 issue)1 and the accompanying editorial2 discuss the attractive and popular notion that the occurrence of fewer infectious diseases during childhood leads to an increase in the incidence of autoimmune diseases. This idea, however, is not soundly based. The apparently increased incidence of autoimmune diseases in industrialized countries may well be attributable to the use of improved diagnostic tools and widespread awareness of these diseases. Large-scale epidemiologic studies have been done only with regard to type 1 diabetes. Moreover, there is evidence that the prevalence of autoimmune diseases in underdeveloped countries is much higher than previously suspected.3 This would be in keeping with the findings in Sardinia, which has the highest prevalence of autoimmune diseases in Europe (similar to that in Finland), despite having, until recently, a high incidence of infectious diseases.4 Finally, the idea that coxsackievirus B can lead to type 1 diabetes is unproved, and the proposal that molecular mimicry (i.e., the presence of viral antigenic determinants that resemble those of the host) may explain the disease in humans has been challenged.5
5 ReferencesSandro Vento, M.D.
Francesca Cainelli, M.D.
University of Verona, 37138 Verona, Italy
ventosandro@yahoo.it 1
Zinkernagel RM . Maternal antibodies, childhood infections, and autoimmune diseases. N Engl J Med 2001;345:1331-1335
Full Text | Web of Science | Medline2
Rosen FS ,Mackay I . The immunology series comes to an end. N Engl J Med 2001;345:1343-1344
Full Text | Web of Science | Medline3
Molokhia M ,McKeigue PM ,Cuadrado M ,Hughes G . Systemic lupus erythematosus in migrants from West Africa compared with Afro-Caribbean people in the UK. Lancet 2001;357:1414-1415
CrossRef | Web of Science | Medline4
Stroffolini T ,Franco E ,Mura I , et al. Age-specific prevalence of hepatitis A virus infection among teenagers in Sardinia. Microbiologica 1991;14:21-24
CrossRef | Web of Science | Medline5
Cainelli F ,Manzaroli D ,Renzini C ,Casali F ,Concia E ,Vento S . Coxsackie B virus-induced autoimmunity to GAD does not lead to type 1 diabetes. Diabetes Care 2000;23:1021-1022
CrossRef | Web of Science | Medline
To the Editor:
Zinkernagel describes the function of maternal antibodies as passive protective mediators that help to generate immunity to invading microbes during early ontogeny. We believe that his view neglects a large body of evidence demonstrating that maternal antibodies also exert a variety of active stimulatory functions on the nascent immune system of the newborn.1,2 Maternal antibodies stimulate T-cell–dependent idiotypic responses that have long-term effects, and they are involved in the establishment of the T-cell repertoire. In addition, in animals, they may substantially enhance immune responses and can transfer a carrier sensitivity (a T-cell–dependent function) from mothers to offspring. In mice, maternal immunization with antigen or anti-idiotypic antibodies induces the production of antigen-reactive IgM antibodies in nonimmunized first-generation offspring. A further argument against the view that maternal antibodies provide only passive protection is the fact that even antibodies that do not react with the antigen — namely, anti-idiotypes — are able to transfer antimicrobial protection from mothers to offspring. Finally, maternally derived IgG antibodies induce long-lasting, allergen-specific suppression of IgE responsiveness in offspring.
We believe that maternal antibodies indeed exert a variety of important active stimulatory functions. These depend on idiotypic interactions among B and T cells, which thus form a physiologically active idiotypic network.3,4
4 ReferencesHilmar Lemke, Ph.D.
Hans Lange, Ph.D.
Christian Albrechts University at Kiel, D-24098 Kiel, Germany
hlemke@biochem.uni-kiel. de 1
Lemke H ,Lange H . Is there a maternally induced immunological imprinting phase à la Konrad Lorenz? Scand J Immunol 1999;50:348-354
CrossRef | Web of Science | Medline2
Idem. Generalization of single immunological experiences by idiotypically mediated clonal connections. Adv Immunol (in press).
3
Jerne NK . The Nobel Lectures in Immunology: the Nobel prize for physiology or medicine, 1984: the generative grammar of the immune system. Scand J Immunol 1993;38:1-9
CrossRef | Web of Science | Medline4
Janeway CA Jr . How the immune system works to protect the host from infection: a personal view. Proc Natl Acad Sci U S A 2001;98:7461-7468
CrossRef | Web of Science | Medline
Author/Editor Response
Dr. Zinkernagel replies:
To the Editor: My article is not comprehensive, and there are many host and environmental factors that influence autoimmune disease.1 Better epidemiologic data in different human populations would clearly be very valuable. I do not want to push the notion that infants with fewer infections have a higher incidence of autoimmune disease than other infants; rather, I would like to argue that infants with insufficient maternally transmitted immunoglobulins, or “clean kids,” either are not sufficiently protected or become infected too late to benefit from maternal antibody protection. I propose that the transfer of maternal antibodies offers a mechanism for “natural,” or “physiologic,” vaccination by attenuating early infections in infants. This mechanism balances decreasing passive maternal protection with active infection and with the maturing immune system of the baby. It explains why vertebrates transfer antibodies to offspring and why emerging infections are generally more severe in the first few generations of new hosts than in subsequent generations.
I agree that mimicry, though possible, is rare and that in many cases, this possibility is supported only by poorly specific assays in vitro. Therefore, the overall role of mimicry in autoimmune disease remains elusive.2 HLA–disease associations have been observed particularly frequently in patients with autoimmunity, which may indicate that chronic or subclinical infections can induce autoimmune disease, even though they are often not recognized as triggering events.3 I propose that during the phase of protection provided by maternal antibodies, favorable, physiologic immunizations of offspring limit early infections, including those that eventually lead to chronic immunopathologic conditions (or autoimmunities). Factors such as increased hygiene and the absence of breast-feeding or the use of breast-feeding for short periods influence the time and dose kinetics of some infections unfavorably, shifting them toward autoimmunity. Such shifts may also be found in developing countries; the increase in the incidence of diabetes in adults between 20 and 40 years of age in southern India is perhaps an example.4
In theory, the postulated active effects of maternal antibodies on the immune system of the offspring by way of idiotypic networks cannot be ruled out. But I am unaware of solid evidence of a survival advantage or disadvantage with respect to infections or autoimmune diseases. Although anti-idiotypic responses may be demonstrated in selected experimental or clinical conditions,5 such evidence does not prove that idiotypic networks have active regulatory roles, either in general or as a result of the influence of maternal antibodies on B-cell or T-cell responses of the offspring. Examination of the cited references suggests that these effects are probably dwarfed in importance by the evolutionary selection pressures exerted by infections that kill offspring early.
5 ReferencesRolf M. Zinkernagel, M.D.
University Hospital, CH-8091 Zurich, Switzerland
rolf.zinkernagel@pty. usz. ch 1
Bach JF . Protective role of infections and vaccinations on autoimmune diseases. J Autoimmun 2001;16:347-353
CrossRef | Web of Science | Medline2
Benoist C ,Mathis D . Autoimmunity provoked by infection: how good is the case for T cell epitope mimicry? Nat Immunol 2001;2:797-801
CrossRef | Web of Science | Medline3
Haddad J ,Deny P ,Munz-Gotheil C , et al. Lymphocytic sialadenitis of Sjogren's syndrome associated with chronic hepatitis C virus liver disease. Lancet 1992;339:321-323
CrossRef | Web of Science | Medline4
Ramachandran A ,Snehalatha C ,Kapur A , et al. High prevalence of diabetes and impaired glucose tolerance in India: National Urban Diabetes Survey. Diabetologia 2001;44:1094-1101
CrossRef | Web of Science | Medline5
Levy R, Miller RA. Therapy of lymphoma directed at idiotypes. In: The lymphomas: current concepts in pathogenesis and management. Journal of the National Cancer Institute monographs. No. 10. Bethesda, Md.: National Cancer Institute, 1990:61-8. (NIH publication no. 90-1930.)
