Modifiers ofthe human mucosal immune system. MAREE GLEESON/ ALLAN W CRIPPS^ and ROBERT L CLANCY'. ' Hunter Immunology^ Unit. Hunter Area ...
Immunology and Cell Biology (1995) 73, 397-404
Modifiers ofthe human mucosal immune system MAREE GLEESON/ ALLAN W CRIPPS^ and ROBERT L CLANCY' ' Hunter Immunology^ Unit. Hunter Area Pathology^ Service. Royal Newcastle Hospital and Faculty of Medicine & Health Sciences. University of Newcastle. Newcastle, New South Wales and-Faculty of Applied Science. University of Canberra, Belconnen, Australian Capital Territory. Australia Summary This review focuses on saliva as a measure of mucosal immunity in man. The review will cover studies of parameters that modify the early ontogeny pattems of mucosal immunity and the impact of infections and physiological variables on the human mucosal immune response. The most significant modifiers of human mucosal immunity are events that occur in the neonatal maturation period and, later in life, the interplay between the immune system and the neuroendocrine systems. IgA antibodies are the predominant isotype involved in the human mucosal immune response and are important for protection at mucosal surfaces. The level of IgA in mucosal secretions is modified by antigenic stimulation as well as by many physiological variables. Studies have also revealed that IgM plays a significant immunoregulatory role at mucosal surfaces, particularly during episodes of infection or stress. The detection patterns of IgD in saliva of neonates suggests a role for IgD in the initial maturation process of mucosal immunity. The role of IgG at mucosal surfaces is unclear and although IgG may play a compensatory role in IgA deficiency, the detection of high levels of IgG in saliva appears to be associated with periods of increased membrane permeability. Key words: albumin., exercise, feeding, immunoglobulins, infection, ontogeny, saliva, stress.
Introduction Scientific studies involving humans have always been complicated by design difficulties and ethical considerations that can often be overcome by using animal models. However, it is important to conduct studies of humans, as a direct correlation between animal models and humans cannot always be inferred. Our research has involved long-term prospective studies ofthe development of mucosal immunity in man and factors that alter the ontogeny pattems. The following section will focus specifically on the use of saliva as a model for monitoring mucosal maturation, immunocompetence and specific immune responses, presenting both the advantages and disadvantages of measurements in saliva.
The saliva model The collection of saliva from the buccal cavity has been used by many researchers ofthe mucosal immune system in man. We have used this model for the measurement of total immunoglobulins.'"-^ and specific antibodies."^ The use of saliva for studies of ontogeny in children satisfied the criteria for: (i) a non-invasive sample collection procedure that was suitable for repeated application; and (ii) a secretion that adequately reflected the mucosal immune system and could be processed in large numbers by appropriate laboratory procedures. Correspondence: Dr Maree Gleeson. Hunter Immunology Unit, Royal Newcastle Hospital. PO Box 664J, Newcastle, NSW 2300, Australia. Received 16 June 1995; accepted 16 June 1995.
The salivary glands have long been recognized as part of the common mucosal immune system.^ However, a criticism ofthe use of saliva to study the mucosal immune system has been that the antibodies in saliva most likely reflect the end result of the mucosal immune response following migration of plasma cells into the salivary gland and may not reflect the initiation phase ofthe response. A second consideration when using saliva is the influence of physiological variations such as flow rate on the concentration of individual proteins in saliva. To adjust for these influences, the levels of albumin or total protein in saliva have been used as a correction factor when necessary. Given these limitations, analysis of salivary immunoglobulins and specific antibodies have provided considerable insight into the development and integrity ofthe mucosal immune system in man. Saliva collections have also been used by researchers for immuno-epidemiological studies of the influence of various parameters on the adult human mucosal immune system. The model has proven particularly suitable for studies where repeated invasive collection techniques are inappropriate and in field situations where sophisticated sample processing has not been available.
Flow rate Increases in flow rate as a result of stimulation have been demonstrated to decrease the concentration of IgA in whole saliva.* The changes in flow rate may account for the higher levels of not only IgA* but all immunoglobulin isotypes and albumin in fasting collections of saliva^ and the lower levels of IgA in saliva of subjects who smoke cigarettes.^ The within-subject variability of immuno-
398
M Gleeson et al.
globulin levels in saliva is also greater in fasting and stimulated collections of saliva than in post-prandial samples collected without stimulation.'^ Differences in collection protocols, including the source of saliva^ often complicate comparisons between studies.
Ontogeny ofthe mucosal immune system There have been very few studies of the development of lymphoid structures ofthe human mucosal immune system, and for ethical reasons all have been performed on post-mortem or post-surgery material. A review by Spencer and MacDonald'° concluded that in the human neonate the cellular components of the gut-associated and bronchus-associated lymphoid tissues were histologically mature at birth and capable of mounting an immune response. Specific studies of the human salivary gland have shown that by 20 weeks gestation the epithelium contains secretory component (SC) and cells expressing cytoplasmic IgA and IgM are present in the mesenchymal tissue." At birth, the mucosal immune system of a healthy neonate is naive, but is rapidly stimulated by bacterial colonization of the gut and extemal surfaces. The human intestinal lamina propria has been shown to be devoid of plasma cells at birth.^ IgM and a few IgG plasma cells appear within a few days of birth and IgA plasma cells appear by day 12 and increase in number to be the predominant isotope by 1 month of age.'^ The study of ontogeny by our research group involved the prospective study of a well-defined community group of healthy infants, recruited at birth from full term healthy pregnancies, and followed for up to 15 years.'-^ This is a unique cohort and the extensive 15 year data base consists of family demographics, family health histories, feeding data for the first 5 years of life on each subject and prospectively documented episodes of atopy, asthma and respiratory and gastrointestinal tract infection. The results of these studies reflect the ontogeny pattems for mucosal immunity in a developed country where health standards are high, and differ from the pattems observed in developing countries.
Salivary IgA Our studies indicated that IgA was rarely found in saliva at birth.' and less than 5% had detectable levels of IgA in the first week of life.' ^ By 1 month of age the reverse situation was observed, with only 3% of the population remaining with no detectable salivary IgA.'^ The median levels of total salivary IgA increased rapidly to peak levels between 4-6 weeks of age followed by a decline to lower levels at 3 months of age.''^ This pattem of appearance of IgA in saliva is consistent with the population of the neonatal intestine with IgA-containing plasma cells'^ and indicated a rapid maturation process of the human mucosal immune system. The reports by other researchers of total IgA and IgA specific antibodies in saliva at birth may reflect differences in study populations, particularly in developing countries where enteric diseases are
endemic.^"* It has also been postulated that the antibodies observed in saliva at birth may reflect an antiidiotype response.'^ During the first year of life, episodes of transient absences of IgA from the saliva collections were noted in 18% ofthe population. A follow up ofthe children, then aged 6-13 years of age, revealed that the transient absences were associated with an increased susceptibility to bronchial hyper-reactivity (BHR) but were not associated with asthma defined as current wheeze."^ The study also indicated a trend towards a negative association with atopy. These results suggested a hypoimmune mucosal immune response in some children during the maturation period. The response of these children to respiratory infection may lead to a persistent or low grade inflammatory response that later in life is reflected by increased BHR, not associated with the clinical symptoms of asthma. The population median for total salivary IgA level remained low and relatively constant throughout the first 4 years of life.'''^ There was an increase in the level of total salivary IgA during the fourth year of life and a dramatic increase in the fifth year'' to levels significantly higher than in adults."^ The level of salivary IgA declined again and remained constant after 7 years of age,** and showed no significant age effect in adults aged between 15-70 years.' The within-subject variability of salivary IgA levels increases with age" and contributing factors in children and adults will be discussed in the following sections. The first 12 months of life appears to be a critical period for the maturation ofthe mucosai immune system. Although the cellular apparatus is in place at birth'- and salivary IgA was detected within 1 month in 97% of the normal population'^ not all infants have the secretory form of IgA in saliva.'^ Despite SC being detected in all saliva samples, approximately one third of children tested in the first 12 months of life had the monomeric form of IgA in some saliva samples.'^ Longitudinal studies indicated that these children switched from having monomeric IgA to secretory IgA in saliva after varying time periods in the first 12 months. The clinical significance of this conversion period is unknown but may highlight a group of children with impaired integrity or delayed maturation of their mucosal immune system. The ontogeny pattems for specific salivary IgA antibodies differ from the pattem for total salivary IgA. Again the source of the study populations give rise to differences between studies. Salivary IgA antibodies to Escherichia coli O antigens remain low in children from developed countries'*-"* and do not parallel the ontogeny pattems for total salivary IgA.'' The development of salivary IgA antibodies to E. coli O antigens'" and Haemophilus influenzae^'^ in countries with lower standards of hygiene, health and socio-economic conditions indicates that antibodies appear in children earlier and at higher levels than in developed countries. Mellander et al.^^ also demonstrated that the level of salivary antibodies were higher in children attending day-care centres, welfare clinics or surgical wards compared to age-matched children in the general community. The differences observed clearly reflect the level of antigenic exposure and persistence of antigenic load in these groups.
399
Human mucosal immunity
Salivary IgM
faces followed by appropriate T cell switch signals for the induction of an IgA response at mucosal surfaces.
IgM is the second major class of secretory immunoglobulins in man and appears to play a significant compensatory role in IgA-deficient adults.^' Salivary IgM is rarely detected in children'^ and adults'' and the ontogeny pattems differ from salivary IgA. IgM was absent from saliva at birth in our study population and appeared in only 15-23% of the neonates between 1 and 6 months of age.'-'^ Salivary IgM did not remain detectable in all samples after the initial period of absence and in individual neonates went through cycles of detectability. The proportion of saliva samples with detectable IgM decreased in the second and third years of life,'^ after which it increased to approximately 20% ofthe population, consistent with observations in adults.^
Salivary IgG and albumin While there are small numbers of IgG producing plasma cells at mucosal surfaces soon after birth'^ and it is possible that local synthesis and secretion can occur, it is unlikely to totally account for the high levels of IgG observed in saliva in 54% of healthy neonates.' '^ There was a strong correlation between the levels of salivary IgG and albumin, and both decreased in parallel over the first 3 months of life.''^ Salivary albumin levels were significantly higher for samples with detectable IgG compared to those with no detectable IgG at all ages, including adults (Fig. 1). Regression models fitted to salivary albumin levels showed that the presence or absence of IgG was a strong predictor of the albumin level. These observations suggest that elevated levels of salivary IgG and albumin reflect periods of increased membrane permeability. Studies of healthy children indicated that in some neonates there was membrane permeability at birth followed by a rapid maturation or closure ofthe epithelium.' '-^ A similar pattem of IgG has been observed in bronchial aspirates from premature infants.'^- further suggesting that the detection of IgG in mucosal secretions reflects periods of increased mucosal membrane permeability. The proportion of samples with detectable IgG fell to almost zero in the first 3 months'^ and remained low for the first year of life (Fig. 2). The detection of IgG in saliva within an individual subject was also cyclical. The frequency of detection increased to adult proportions by 4
The pattems of neonatal development showed a strong association between the concentrations of IgA and IgM in saliva.'^ Median IgA levels were significantly higher in infants with detectable salivary IgM than in those with no detectable IgM." In the neonate, when salivary IgA was not detected, salivary IgM was almost always undetectable.' ^ The compensation by salivar>' IgM observed in IgA-deficient adults^' was not evident in the neonatal period, which may reflect immaturity ofthe mucosal immune system at birth. The correlation between concentrations of salivary IgA and IgM suggested that their appearance in mucosal secretions and the initial peak level observed shortly after birth were stimulated by exposure to novel antigens. The subsequent fall in salivary IgA and IgM levels in the first 3 months suggests normal Tcell control mechanisms operating at the mucosal sur-
70 -1
EO •
50 -
O)
uo -
30 -
20 -
10 •
20
—I60
—I—
ao
—I— 100
200
>100
600
BOO
1000
I IOO
UOO
Adult
Age (days) Figure I Geometric means and 95% confidence intervals of albumin for saliva samples with detectable IgG (•) and for samples with no detectable IgG (O). The age groups range from birth to 4 years of age and for an adult group aged 15-65 years.
400
M Gleeson et al.
0.-6 -,
0.5
-
0.4
-
/< 1
CO
o a. ^ o c
o
O Q. O
>
4
0.3 i N
-
t 0.2
1
