- Email: [email protected]
Class-specific antibodies to Streptococcus mutans in human serum, saliva and breast milk
Journal oflmmunologicalMethods, 87 (1986)103-108 Elsevier
103
JIM03851
Class-specific antibodies to Streptococcus mutans in human serum, saliva and breast milk M.J. A l d r e d *, W . G . W a d e , D.R. Llewelyn I a n d D . M . W a l k e r Department of Oral Medicine and Oral Pathology and I Department of Children's and Preventive Dentistry, Dental School, University of Wales College of Medicine, Heath Park, Cardiff CF4 4XY, Wales, U.K. (Accepted 25 November 1985)
Previous techniques used for the detection and quantitation of antibodies in body fluids may be inappropriate where only small volumes are available, or may not be sensitive enough to detect low levels of specific antibodies. An indirect ELISA technique has successfully been employed to estimate class-specific antibody levels to Streptococcus m u t a n s in serum and secretions in a group of mothers and their neonates, and an attempt has been made to relate such levels to the presence or absence of active caries in the mothers. A high maternal serum IgG antibody level appears to exert a protective effect against dental caries. Antibody levels in maternal saliva and colostrum/breast milk showed no differences between the 2 groups. The presence of active caries in mothers was associated with an elevated IgA antibody level in neonatal saliva. Although ELISA permitted the detection of low levels of antibody in the small volumes of neonatal saliva collected, a further increase in sensitivity and specificity of the assay would be advantageous. Key words: Indirect ELISA; Streptococcus mutans; Dental caries," Serum," Secretions
Introduction Considerable evidence incriminates Streptococcus m u t a n s as the most important organism in the initiation of dental caries (McGhee and Michalek, 1981). High levels of serum IgG antibody to S. m u t a n s are associated with caries resistance in man (Challacombe, 1980; Challacombe et al., 1984) * Address correspondence to: M.J. Aldred, Department of Oral Medicine and Oral Pathology, Dental School, University of Wales College of Medicine, Heath Park, Cardiff CF4 4XY, Wales, U.K. Abbreviations: S. mutans, Streptc~occus mutans; DMF, decayed, missing, filled (teeth); TYCSB, tryptone, yeast extract, cystine agar incorporating added sucrose and bacitracin; PBS, phosphate-buffered saline; Ig, immunoglobulin; ELISA, enzyme-linked immunosorbent assay; BSA, bovine serum albumin; CFU, colony-forming units.
and in immunised animals (Lehner et al., 1977). Breast feeding protects infants against infection, particularly gastroenteritis (Totterdell et al., 1976) and specific antibodies in breast milk appear responsible (Glass et al., 1983). Breast fed infants are less caries susceptible than are bottle fed children (Tank and Storvick, 1965) and secretory antibodies to S. m u t a n s in colostrum (Arnold et al., 1976; Eggert and Gurner, 1984) and in neonatal saliva (Bammann and Gibbons, 1979) may be responsible. Alternatively, cross-placental transfer of antibodies from the maternal bloodstream might be more important in conferring passive immunity upon an infant. In this study using an indirect ELISA technique, antibody levels to S. m u t a n s have been assayed in maternal breast milk or colostrum, serum and saliva and in neonatal saliva and cord
0022-1759/86/$03.50 ~ 1986 Elsevier Science Publishers B.V. (Biomedical Division)
104
blood serum. Since the presence of active caries influences antibody levels to S. mutans (Challacombe, 1980) the mothers in this study were subdivided according to whether they had active caries or treated caries. In addition, the effect of breast feeding on antibody levels to S. mutans in neonatal saliva has been considered.
Subjects, materials and methods
Subjects 79 mothers and their neonates were studied within the first 9 days post partum, with the mother's informed consent. The subjects were divided into a treated caries group and an active caries group according to the absence or presence of 1 or more carious cavities detected using a dental mirror and sharp probe, excluding those cavities which were considered as arrested caries. Radiographs were not employed. The decayed, missing, filled ( D M F ) index (W.H.O., 1962) was recorded, excluding third molars from the assessment, to give a measure of previous caries experience. Collection and preparation of samples Maternal and cord blood serum were obtained by standard methods. Colostrum or breast milk was obtained by manual expression and the fatty layer removed and discarded after centrifugation. Whole unstimulated saliva was collected from the mothers by spitting and from the babies via a premature infant feeding tube (H.G. Wallace) connected to a sterile fluid trap, with negative pressure applied using a fixed wall suction unit. Solid material was removed by centrifugation and the supernatant retained for use in the assay. An aliquot from the saliva of 44 of the mothers was diluted in phosphate-buffered saline (PBS) (Oxoid) and plated out onto tryptone, yeast extract, cystine agar incorporating added sucrose and bacitracin (TYCSB) (Wade et al., 1983) - a selective medium for S. mutans - to obtain an estimate of the salivary S. mutans count. All samples were stored at - 2 0 ° C until required. EL1SA An indirect ELISA (Voller and Bidwell, 1975)
was used to assay IgA, IgG and IgM antibody levels to S. mutans. Whole cells of S. mutans D283 (serotype c) suspended at an optical density of 1.0 U at 540 nm in carbonate-bicarbonate buffer pH 9.6 were used to coat the wells of polystyrene microtitre plates (Flow Laboratories) 200 /~l per well and incubated at 4°C overnight. After three 5 min washes with tap water containing 0.05% Tween 20 (Koch-Light), serum diluted 1/200 or secretions at a dilution of 1/10, each in P B S / I % bovine serum albumin (BSA) (Armour Pharmaceuticals)/0.05% Tween 20 was incubated in the wells for 2 h at room temperature, and after further washing goat anti-human IgA, IgG or IgM conjugated to alkaline phosphatase (Sigma) each diluted 1/1000 in the same diluent as for serum was incubated under the same conditions. Following a further wash cycle, the substrate, pnitrophenyl phosphate (Sigma) in diethanolamine (Sigma) buffer p H 9.8 was added to each of the wells. The plate was read (Titertek Multiskan, Flow Laboratories) when a pooled reference positive serum reached an absorbance at 405 nm of 0.5 U for serum and 1.0 U for milk or saliva. All samples were assayed in duplicate and the mean absorbance calculated. Desorption of whole cells from the solid phase Preliminary experiments were conducted in which the antigen solution was harvested after incubation in the wells, as was the fluid from the subsequent nine washes. The absorbance at 540 nm was recorded as described above after resuspension to the original volume, and the absorbance compared with the cell density as determined by direct counting. Sensitivity and specificity of the assay Serial dilutions of up to 1 : 2 5 6 0 0 of human serum, colostrum/breast milk and saliva were used in the assay described above, incorporating wells containing no antigen a n d / o r no sample as controls. The absorbance was read when a reference positive serum dilution of 1:200 reached an absorbance of 0.5 U a t 405 nm. Specificity was determined by preabsorption of samples at the appropriate working dilution with equal volumes of packed whole cells of S. mutans D283 or C. albicans N C P F 3153, at 4°C overnight, following
105 1.0-
0.8
p<0.05
0.8
0.6
~7 P
0.6
iilil
•
Active Caries
........
[]
T r e a t e d Caries
i!E~
•
Active Caries
[] ]
T r e a t e d Caries
Background
0.4
:::: w
0.4
0.2
0
n 23 48 IgA o
n 23 56 IgA
23 56 IgG
23 48 IgG
23 48 IgM
Fig. 2. Geometric mean class-specific antibody levels to S. in maternal saliva in active caries and treated caries groups.
23 56 IgM
mutans
Fig. 1. Geometric mean class-specific antibody levels to S. in maternal serum in active caries and treated caries groups. Note that IgA and IgG antibody levels increased with age and the treated caries group was older than the active caries group. After adjustment for this age difference only IgG antibody levels remained significantly different.
mutans
which the supernatant was retrieved after centrifugation at 4000 x g for 10 min and used in the ELISA.
Statistical analys& Logarithmic transformation of all antibody levels was performed before further analysis. This effected a normal distribution with respect to maternal antibody levels to which Student's t-test was applied, but the non-parametric Mann-Whitney U-test was required for the cord blood serum and neonatal salivary antibody levels.
Results
Desorption of whole cells from the solid phase After the initial incubation of antigen, the fluid retrieved from the wells registered an absorbance of 0.475 U at 540 nm, and after addition of the cells lost during washing this rose to 0.501 U, indicating that approximately 50% of the cells remained attached to the plate. Comparison of absorbance with cell numbers confirmed a linear correlation, with 9 x 105 cells/ml in the original antigen preparation, and 4.55 x 105 cells/ml in the retrieved fluid.
Sensitivity and specificity of the assay Individual serum samples at a dilution of up to 1 : 25 600, c o l o s t r u m / b r e a s t milk at a dilution of up to 1 : 5120 and adult saliva at a dilution of up
TABLE I DETAILS O F M O T H E R S A N D BABIES IN ACTIVE CARIES A N D T R E A T E D CARIES G R O U P S Group
Active caries Treated caries * P < 0.05.
n
23 56
Maternal mean age (years)
Age range (years)
Mean D M F index
24.8 * 27.3 *
17-32 17-39
12.9 14.7
Mean S. mutans
% Breast feeding
Geometric mean age of babies (days)
48 70
2.8 3.2
CFU/ml 2.6 X 105 8.9 × 104
106 1.2
0.6
1
Active Caries p<0.05
1.0
0.4
0.8
[]
Treated Caries
]
Background
0.2
Active Caries E u~ 0.6
[]
Treated Caries
[]
Background
0 n
LU 0.4
19 39
19 39
IgA
IgG
18 38 IgM
Fig. 4. Median class-specific antibody levels to S. mutans in neonatal saliva in active caries and treated caries groups.
0.2
n 11 33
IgA
11 33
11 33
IgG
IgM
Comparison of clinical data in active caries and treated caries groups The mothers with treated caries were significantly older than those with active caries (Table I). N o significant differences between the treated caries and active caries groups were found in the maternal D M F index, salivary count of S. m u t a n s , in the proportion of mothers breast feeding, or in the m e a n age of their neonates. In the active caries group there was a total of 37 cavities (median 1 cavity, range 1 - 5 cavities).
Fig. 3. Geometric mean class-specific antibody levels to S. rnutans in colostrum/breast milk in active caries and treated caries groups.
to 1 : 1 2 8 0 , showed absorbances above the background level, while preabsorption with the hom o l o g o u s organism markedly reduced the antib o d y level recorded at the appropriate working dilution, but only a negligible reduction in level was observed after preabsorption with C. a l b i c a n s
Comparison of antibody levels to S. mutans in serum and secretions in active caries and treated caries groups Serum antibody levels of all classes assayed were significantly lower in mothers with active
N C P F 3153.
•
Active Caries
[]
Treated Caries
B r e a s t Fed E
Background
]
0.6
Bottle
Fed
0.4
u~
0.2
0
-
n
8
27
IgA
8
-
27
8 26
11 12
11 12
10 12
IgG
IgM
IgA
IgG
IgM
Fig. 5. Median class-specific antibody levels to S. mutans in neonatal saliva in active caries and treated caries groups according to mode of feeding.
107
caries compared to those in the treated caries group (Fig. 1). However, in the group as a whole, maternal serum IgA and IgG antibody levels increased with maternal age ( P < 0.05 and 0.1 > P > 0.05 respectively, data not shown) and after adjustment for age only IgG antibody levels remained significantly lower in the active caries group compared with the treated caries group (t = 2.14, P < 0.05). There was no significant difference in the levels of class-specific antibodies in maternal saliva (Fig. 2) or in colostrum/breast milk in the 2 groups (Fig. 3). Median cord blood serum IgG levels were significantly lower in the AC group than in the treated caries group, ( P < 0.05), but this was not maintained when the maternal age difference between the 2 groups was taken into account. Elevated IgA levels to S. mutans were detected in the saliva of babies of mothers with active caries compared to those of babies of mothers with treated caries (Fig. 4). When these babies were subdivided into breast and bottle fed groups (Fig. 5), levels were still higher in the active caries group in both breast and bottle fed infants, although the differences were not statistically significant.
Discussion
The desorption of whole cells from the polystyrene plates recorded in this study is of the same order as the results of Elder et al. (1982), who used radiolabelling to assess the binding of S. sanguis to the solid phase. The use of linking agents has been described to enhance binding of whole cells to the plastic support (Challacombe et al., 1984). Serum and salivary antibodies to S. mutans have been measured previously by agglutination (Kennedy et al., 1968; Challacombe, 1978, 1980; Huis in 't Veld et al., 1978), indirect immunofluorescence (Challacombe, 1980) or radioimmunoassay (Czerkinsky et al., 1983; Challacombe et al., 1984). The ELISA method used in the present investigation and by others (Huis in 't Veld et al., 1978; Bolton and Hlava, 1982; Lehtonen et al., 1984; Aaltonen et al., 1985) has the advantage of being relatively inexpensive and sensitive enough to overcome the problems of de-
tecting low levels of specific antibody in the small sample volumes available, e.g., in neonatal saliva. The ELISA also avoided the difficulties inherent in the estimation of salivary antibodies by agglutination techniques since saliva contains nonimmunoglobulin agglutinating factors (Magnusson and Ericson, 1977). The difference in age between the AC and TC groups probably reflects the random selection of subjects employed, whereas other studies have used a much narrower age range (Challacombe et al., 1973; Challacombe and Lehner, 1976). The serum antibody levels after adjustment for age in the active caries group compared to those of the treated caries subjects of similar D M F suggested that a low serum IgG antibody level to S. mutans was associated with active caries. Our results add further conflicting evidence to the reports regarding natural immunity to dental caries in humans. Previous studies have suggested that serum antibody levels are elevated as a result of active caries compared with levels in a control group with similar previous caries experience but no untreated caries (Challacombe, 1980) although this may depend upon the antigen employed (Challacombe et al., 1984). Further confusion may arise since some reports have combined subjects with treated caries and with no previous caries (caries-resistant) in comparing antibody levels to whole cells or cell extracts (Challacombe et al., 1973; Challacombe and Lehner, 1976), and others have allocated individuals with increased previous caries experience into an 'active' caries group (Huis in 't Veld et al., 1978). Since all of the mothers in this study had some previous caries experience our results cannot be directly compared with those of Kennedy et al. (1968) or Huis in 't Veld et al. (1978). The duration of the carious lesions in the active caries subjects studied here was not known, nor was the time since the last treatment recorded for the treated caries group, and in this respect changes in antibody levels related to the development or treatment of lesions may be relevant (Challacombe and Lehner, 1976; Challacombe, 1980). No difference in maternal salivary antibody levels to S. mutans in the 2 groups was apparent, in accord with the study of Huis in 't Veld et al. (1978), who examined parotid saliva, although in
108
some series lower IgA levels have been found in subjects with active caries in parotid saliva (Challacombe, 1978, 1980) or mixed saliva (Bolton and Hlava, 1982). Cord blood serum antibody levels to S. m u t a n s did not appear to be significantly influenced by the caries status of the mother when the effect of maternal age on antibody levels was taken into account. The mechanism responsible for the significantly higher median IgA level in neonatal saliva in the active caries group compared to the treated caries group is unclear. Since this trend was seen in both breast and bottle fed babies it cannot be related to passive immunity derived by breast feeding and some factor, possibly transferred in cord blood as suggested by Ivanyi and Lehner (1978) may be responsible for stimulation of lymphocytes to produce immunoglobulins. Since S. m u t a n s does not in general establish itself in the mouth until teeth have erupted (Berkowitz et al., 1975, 1980; Catalanotto et al.~ 1975), usually after breast feeding has been discontinued (Martin, 1978) the mode of infant feeding may not directly influence the colonisation of the infant's mouth and other factors such as diet may play a more significant role. Since the amount of saliva obtainable from the infants was limited, and antibody levels were in general rather low, it would be of considerable advantage if the sensitivity could be increased still more. Furthermore, the use of purified antigens rather than whole cells (Russell et al., 1980~ 1982; Lehner et al., 1985) should prove more specific than whole bacterial cells in the assay.
Acknowledgements The authors wish to thank Dr. R.G. Newcombe for advice regarding statistical analysis of the data. Supported by a grant from the Welsh Scheme for the Development of Health and Social Research.
References Aaltonen, A.S., J. Tenovuo, O.-P. Lehtonen, R. Saksala and O. Meurman, 1985, Arch. Oral Biol. 30, 331. Arnold, R.R., J. Mestecky and J.R. McGhee, 1976, Infect. Immun. 14, 355.
Bammann, L.L. and R.J. Gibbons, 1979, J. Clin. Microbiol. 10, 538. Berkowitz, R.J., H.V. Jordan and G. White, 1975, Arch. Oral Biol. 20, 171. Berkowitz, R.J., J. Turner and P. Green, 1980. Arch. Oral Biol. 25, 221. Bolton, R.W. and G . L Hlava, 1982, J. Dent. Res. 61, 1225. Catalanotto, F.A., I.L. Shklair and H.J. Keene. 1975, J. Am. Dent. Assoc. 91,606. Challacombe, S.J., 1978. Adv. Exp. Med. Biol. 107, 355. Challacombe, S.J., 1980, Arch. Oral Biol. 25,495. Challacombe, S.J. and T. Lehner, 1976, J. Dent. Res. 55, 139. Challacombe, S.J., B. Guggenheim and T. Lehner, 1973, Arch. Oral Biol. 18, 657. Challacombe, S.J., E.A. Bergmeier and A.S. Rees, 1984. Arch. Oral Biol. 29, 179. Czerkinsky, C., A.S. Rees, L A . Bergmeier and S.J. Challacombe, 1983, Clin. Exp. lmmunol. 53, 192. Eggert, F.M. and B.W. Gurner, 1984, Infect. Immun. 44, 660. Elder, B.L., D.K. Boraker and P.M. Fives-Taylor, 1982, J. Clin. Microbiol. 16, 141. Glass, R.I., A.-M. Svennerholm, B.J. Stoll, M.R Khan, K.M.B. Hossain, M.I. Huq and J. Holmgren, 1983, N. Engl. J. Med. 308, 1389. Huis in "t Veld, J., D. Bannet, W. Van Palenstein Helderman, P.S. Camargo and O. Backer-Dirks, 1978, Adv. Exp. Med. Biol. 107, 369. lvanyi, L. and T. Lehner, 1978, Arch. Oral Biol. 23, 851. Kennedy, A.E., I.L. Shklair, J.A. Hayashi and A.N. Bahn, 1968, Arch. Oral Biol. 13, 1275. KOhler, B., B.-M. Pettersson and D. Bratthall, 1981, Scand. J. Dent. Res. 89, 19. Lehner, T., J. Caldwell and S.J. Challacombe, 1977, Arch. Oral Biol. 27, 393. Lehner, T., J. Caldwell and A.S.M. Giassudin, 1985, Arch. Oral Biol. 30, 207. Lehtonen, O.-P.J., E.M. Grahn, T.H. Stahlberg and L.A. Laitinen, 1984, Infect. lmmun. 43, 308. McGhee, J.R. and S.M. Michalek, 1981, Ann. Rev. Microbiol. 35, 595. Magnusson, I. and T. Ericson, 1977, Caries Res. 11, 109. Martin, J., 1978, Infant Feeding 1975: Attitudes and Practice in England and Wales (H.M.S.O., London). Michael, J.G., R. Ringenback and S. Hottenstein, 1971, J. Infect. Dis. 124, 445. Ogra, S.S. and P.L. Ogra, 1978, J. Pediatr. 92, 546. Russell, M.W., L.A. Bergmeier, E.D. Zanders and T. Lehner, 1980, Infect. lmmun. 28, 486. Russell, R.R.B., D. Beighton and B. Cohen, 1982, Br. Dent. J. 152, 81. Tank, G. and C.A. Storvick, 1965, J. Am. Dent. Assoc. 70, 394. Totterdell, B.M., I.E. Chrystie and J.E. Banatvala, 1976, Arch. Dis. Child. 51,924. Voller, A. and D.E. Bidwell, 1975, Br. J. Exp. Pathol. 56, 338. Wade, W.G., D.R. Llewelyn, D.M. Walker and M.J. Aldred, 1983, J. Dent. Res. 62, 420. World Health Organisation, 1962, Technical Report Series No. 242, Standardisation of reporting on dental diseases and conditions, Report of an expert committee on dental health (W.H.O., Geneva).
103
JIM03851
Class-specific antibodies to Streptococcus mutans in human serum, saliva and breast milk M.J. A l d r e d *, W . G . W a d e , D.R. Llewelyn I a n d D . M . W a l k e r Department of Oral Medicine and Oral Pathology and I Department of Children's and Preventive Dentistry, Dental School, University of Wales College of Medicine, Heath Park, Cardiff CF4 4XY, Wales, U.K. (Accepted 25 November 1985)
Previous techniques used for the detection and quantitation of antibodies in body fluids may be inappropriate where only small volumes are available, or may not be sensitive enough to detect low levels of specific antibodies. An indirect ELISA technique has successfully been employed to estimate class-specific antibody levels to Streptococcus m u t a n s in serum and secretions in a group of mothers and their neonates, and an attempt has been made to relate such levels to the presence or absence of active caries in the mothers. A high maternal serum IgG antibody level appears to exert a protective effect against dental caries. Antibody levels in maternal saliva and colostrum/breast milk showed no differences between the 2 groups. The presence of active caries in mothers was associated with an elevated IgA antibody level in neonatal saliva. Although ELISA permitted the detection of low levels of antibody in the small volumes of neonatal saliva collected, a further increase in sensitivity and specificity of the assay would be advantageous. Key words: Indirect ELISA; Streptococcus mutans; Dental caries," Serum," Secretions
Introduction Considerable evidence incriminates Streptococcus m u t a n s as the most important organism in the initiation of dental caries (McGhee and Michalek, 1981). High levels of serum IgG antibody to S. m u t a n s are associated with caries resistance in man (Challacombe, 1980; Challacombe et al., 1984) * Address correspondence to: M.J. Aldred, Department of Oral Medicine and Oral Pathology, Dental School, University of Wales College of Medicine, Heath Park, Cardiff CF4 4XY, Wales, U.K. Abbreviations: S. mutans, Streptc~occus mutans; DMF, decayed, missing, filled (teeth); TYCSB, tryptone, yeast extract, cystine agar incorporating added sucrose and bacitracin; PBS, phosphate-buffered saline; Ig, immunoglobulin; ELISA, enzyme-linked immunosorbent assay; BSA, bovine serum albumin; CFU, colony-forming units.
and in immunised animals (Lehner et al., 1977). Breast feeding protects infants against infection, particularly gastroenteritis (Totterdell et al., 1976) and specific antibodies in breast milk appear responsible (Glass et al., 1983). Breast fed infants are less caries susceptible than are bottle fed children (Tank and Storvick, 1965) and secretory antibodies to S. m u t a n s in colostrum (Arnold et al., 1976; Eggert and Gurner, 1984) and in neonatal saliva (Bammann and Gibbons, 1979) may be responsible. Alternatively, cross-placental transfer of antibodies from the maternal bloodstream might be more important in conferring passive immunity upon an infant. In this study using an indirect ELISA technique, antibody levels to S. m u t a n s have been assayed in maternal breast milk or colostrum, serum and saliva and in neonatal saliva and cord
0022-1759/86/$03.50 ~ 1986 Elsevier Science Publishers B.V. (Biomedical Division)
104
blood serum. Since the presence of active caries influences antibody levels to S. mutans (Challacombe, 1980) the mothers in this study were subdivided according to whether they had active caries or treated caries. In addition, the effect of breast feeding on antibody levels to S. mutans in neonatal saliva has been considered.
Subjects, materials and methods
Subjects 79 mothers and their neonates were studied within the first 9 days post partum, with the mother's informed consent. The subjects were divided into a treated caries group and an active caries group according to the absence or presence of 1 or more carious cavities detected using a dental mirror and sharp probe, excluding those cavities which were considered as arrested caries. Radiographs were not employed. The decayed, missing, filled ( D M F ) index (W.H.O., 1962) was recorded, excluding third molars from the assessment, to give a measure of previous caries experience. Collection and preparation of samples Maternal and cord blood serum were obtained by standard methods. Colostrum or breast milk was obtained by manual expression and the fatty layer removed and discarded after centrifugation. Whole unstimulated saliva was collected from the mothers by spitting and from the babies via a premature infant feeding tube (H.G. Wallace) connected to a sterile fluid trap, with negative pressure applied using a fixed wall suction unit. Solid material was removed by centrifugation and the supernatant retained for use in the assay. An aliquot from the saliva of 44 of the mothers was diluted in phosphate-buffered saline (PBS) (Oxoid) and plated out onto tryptone, yeast extract, cystine agar incorporating added sucrose and bacitracin (TYCSB) (Wade et al., 1983) - a selective medium for S. mutans - to obtain an estimate of the salivary S. mutans count. All samples were stored at - 2 0 ° C until required. EL1SA An indirect ELISA (Voller and Bidwell, 1975)
was used to assay IgA, IgG and IgM antibody levels to S. mutans. Whole cells of S. mutans D283 (serotype c) suspended at an optical density of 1.0 U at 540 nm in carbonate-bicarbonate buffer pH 9.6 were used to coat the wells of polystyrene microtitre plates (Flow Laboratories) 200 /~l per well and incubated at 4°C overnight. After three 5 min washes with tap water containing 0.05% Tween 20 (Koch-Light), serum diluted 1/200 or secretions at a dilution of 1/10, each in P B S / I % bovine serum albumin (BSA) (Armour Pharmaceuticals)/0.05% Tween 20 was incubated in the wells for 2 h at room temperature, and after further washing goat anti-human IgA, IgG or IgM conjugated to alkaline phosphatase (Sigma) each diluted 1/1000 in the same diluent as for serum was incubated under the same conditions. Following a further wash cycle, the substrate, pnitrophenyl phosphate (Sigma) in diethanolamine (Sigma) buffer p H 9.8 was added to each of the wells. The plate was read (Titertek Multiskan, Flow Laboratories) when a pooled reference positive serum reached an absorbance at 405 nm of 0.5 U for serum and 1.0 U for milk or saliva. All samples were assayed in duplicate and the mean absorbance calculated. Desorption of whole cells from the solid phase Preliminary experiments were conducted in which the antigen solution was harvested after incubation in the wells, as was the fluid from the subsequent nine washes. The absorbance at 540 nm was recorded as described above after resuspension to the original volume, and the absorbance compared with the cell density as determined by direct counting. Sensitivity and specificity of the assay Serial dilutions of up to 1 : 2 5 6 0 0 of human serum, colostrum/breast milk and saliva were used in the assay described above, incorporating wells containing no antigen a n d / o r no sample as controls. The absorbance was read when a reference positive serum dilution of 1:200 reached an absorbance of 0.5 U a t 405 nm. Specificity was determined by preabsorption of samples at the appropriate working dilution with equal volumes of packed whole cells of S. mutans D283 or C. albicans N C P F 3153, at 4°C overnight, following
105 1.0-
0.8
p<0.05
0.8
0.6
~7 P
0.6
iilil
•
Active Caries
........
[]
T r e a t e d Caries
i!E~
•
Active Caries
[] ]
T r e a t e d Caries
Background
0.4
:::: w
0.4
0.2
0
n 23 48 IgA o
n 23 56 IgA
23 56 IgG
23 48 IgG
23 48 IgM
Fig. 2. Geometric mean class-specific antibody levels to S. in maternal saliva in active caries and treated caries groups.
23 56 IgM
mutans
Fig. 1. Geometric mean class-specific antibody levels to S. in maternal serum in active caries and treated caries groups. Note that IgA and IgG antibody levels increased with age and the treated caries group was older than the active caries group. After adjustment for this age difference only IgG antibody levels remained significantly different.
mutans
which the supernatant was retrieved after centrifugation at 4000 x g for 10 min and used in the ELISA.
Statistical analys& Logarithmic transformation of all antibody levels was performed before further analysis. This effected a normal distribution with respect to maternal antibody levels to which Student's t-test was applied, but the non-parametric Mann-Whitney U-test was required for the cord blood serum and neonatal salivary antibody levels.
Results
Desorption of whole cells from the solid phase After the initial incubation of antigen, the fluid retrieved from the wells registered an absorbance of 0.475 U at 540 nm, and after addition of the cells lost during washing this rose to 0.501 U, indicating that approximately 50% of the cells remained attached to the plate. Comparison of absorbance with cell numbers confirmed a linear correlation, with 9 x 105 cells/ml in the original antigen preparation, and 4.55 x 105 cells/ml in the retrieved fluid.
Sensitivity and specificity of the assay Individual serum samples at a dilution of up to 1 : 25 600, c o l o s t r u m / b r e a s t milk at a dilution of up to 1 : 5120 and adult saliva at a dilution of up
TABLE I DETAILS O F M O T H E R S A N D BABIES IN ACTIVE CARIES A N D T R E A T E D CARIES G R O U P S Group
Active caries Treated caries * P < 0.05.
n
23 56
Maternal mean age (years)
Age range (years)
Mean D M F index
24.8 * 27.3 *
17-32 17-39
12.9 14.7
Mean S. mutans
% Breast feeding
Geometric mean age of babies (days)
48 70
2.8 3.2
CFU/ml 2.6 X 105 8.9 × 104
106 1.2
0.6
1
Active Caries p<0.05
1.0
0.4
0.8
[]
Treated Caries
]
Background
0.2
Active Caries E u~ 0.6
[]
Treated Caries
[]
Background
0 n
LU 0.4
19 39
19 39
IgA
IgG
18 38 IgM
Fig. 4. Median class-specific antibody levels to S. mutans in neonatal saliva in active caries and treated caries groups.
0.2
n 11 33
IgA
11 33
11 33
IgG
IgM
Comparison of clinical data in active caries and treated caries groups The mothers with treated caries were significantly older than those with active caries (Table I). N o significant differences between the treated caries and active caries groups were found in the maternal D M F index, salivary count of S. m u t a n s , in the proportion of mothers breast feeding, or in the m e a n age of their neonates. In the active caries group there was a total of 37 cavities (median 1 cavity, range 1 - 5 cavities).
Fig. 3. Geometric mean class-specific antibody levels to S. rnutans in colostrum/breast milk in active caries and treated caries groups.
to 1 : 1 2 8 0 , showed absorbances above the background level, while preabsorption with the hom o l o g o u s organism markedly reduced the antib o d y level recorded at the appropriate working dilution, but only a negligible reduction in level was observed after preabsorption with C. a l b i c a n s
Comparison of antibody levels to S. mutans in serum and secretions in active caries and treated caries groups Serum antibody levels of all classes assayed were significantly lower in mothers with active
N C P F 3153.
•
Active Caries
[]
Treated Caries
B r e a s t Fed E
Background
]
0.6
Bottle
Fed
0.4
u~
0.2
0
-
n
8
27
IgA
8
-
27
8 26
11 12
11 12
10 12
IgG
IgM
IgA
IgG
IgM
Fig. 5. Median class-specific antibody levels to S. mutans in neonatal saliva in active caries and treated caries groups according to mode of feeding.
107
caries compared to those in the treated caries group (Fig. 1). However, in the group as a whole, maternal serum IgA and IgG antibody levels increased with maternal age ( P < 0.05 and 0.1 > P > 0.05 respectively, data not shown) and after adjustment for age only IgG antibody levels remained significantly lower in the active caries group compared with the treated caries group (t = 2.14, P < 0.05). There was no significant difference in the levels of class-specific antibodies in maternal saliva (Fig. 2) or in colostrum/breast milk in the 2 groups (Fig. 3). Median cord blood serum IgG levels were significantly lower in the AC group than in the treated caries group, ( P < 0.05), but this was not maintained when the maternal age difference between the 2 groups was taken into account. Elevated IgA levels to S. mutans were detected in the saliva of babies of mothers with active caries compared to those of babies of mothers with treated caries (Fig. 4). When these babies were subdivided into breast and bottle fed groups (Fig. 5), levels were still higher in the active caries group in both breast and bottle fed infants, although the differences were not statistically significant.
Discussion
The desorption of whole cells from the polystyrene plates recorded in this study is of the same order as the results of Elder et al. (1982), who used radiolabelling to assess the binding of S. sanguis to the solid phase. The use of linking agents has been described to enhance binding of whole cells to the plastic support (Challacombe et al., 1984). Serum and salivary antibodies to S. mutans have been measured previously by agglutination (Kennedy et al., 1968; Challacombe, 1978, 1980; Huis in 't Veld et al., 1978), indirect immunofluorescence (Challacombe, 1980) or radioimmunoassay (Czerkinsky et al., 1983; Challacombe et al., 1984). The ELISA method used in the present investigation and by others (Huis in 't Veld et al., 1978; Bolton and Hlava, 1982; Lehtonen et al., 1984; Aaltonen et al., 1985) has the advantage of being relatively inexpensive and sensitive enough to overcome the problems of de-
tecting low levels of specific antibody in the small sample volumes available, e.g., in neonatal saliva. The ELISA also avoided the difficulties inherent in the estimation of salivary antibodies by agglutination techniques since saliva contains nonimmunoglobulin agglutinating factors (Magnusson and Ericson, 1977). The difference in age between the AC and TC groups probably reflects the random selection of subjects employed, whereas other studies have used a much narrower age range (Challacombe et al., 1973; Challacombe and Lehner, 1976). The serum antibody levels after adjustment for age in the active caries group compared to those of the treated caries subjects of similar D M F suggested that a low serum IgG antibody level to S. mutans was associated with active caries. Our results add further conflicting evidence to the reports regarding natural immunity to dental caries in humans. Previous studies have suggested that serum antibody levels are elevated as a result of active caries compared with levels in a control group with similar previous caries experience but no untreated caries (Challacombe, 1980) although this may depend upon the antigen employed (Challacombe et al., 1984). Further confusion may arise since some reports have combined subjects with treated caries and with no previous caries (caries-resistant) in comparing antibody levels to whole cells or cell extracts (Challacombe et al., 1973; Challacombe and Lehner, 1976), and others have allocated individuals with increased previous caries experience into an 'active' caries group (Huis in 't Veld et al., 1978). Since all of the mothers in this study had some previous caries experience our results cannot be directly compared with those of Kennedy et al. (1968) or Huis in 't Veld et al. (1978). The duration of the carious lesions in the active caries subjects studied here was not known, nor was the time since the last treatment recorded for the treated caries group, and in this respect changes in antibody levels related to the development or treatment of lesions may be relevant (Challacombe and Lehner, 1976; Challacombe, 1980). No difference in maternal salivary antibody levels to S. mutans in the 2 groups was apparent, in accord with the study of Huis in 't Veld et al. (1978), who examined parotid saliva, although in
108
some series lower IgA levels have been found in subjects with active caries in parotid saliva (Challacombe, 1978, 1980) or mixed saliva (Bolton and Hlava, 1982). Cord blood serum antibody levels to S. m u t a n s did not appear to be significantly influenced by the caries status of the mother when the effect of maternal age on antibody levels was taken into account. The mechanism responsible for the significantly higher median IgA level in neonatal saliva in the active caries group compared to the treated caries group is unclear. Since this trend was seen in both breast and bottle fed babies it cannot be related to passive immunity derived by breast feeding and some factor, possibly transferred in cord blood as suggested by Ivanyi and Lehner (1978) may be responsible for stimulation of lymphocytes to produce immunoglobulins. Since S. m u t a n s does not in general establish itself in the mouth until teeth have erupted (Berkowitz et al., 1975, 1980; Catalanotto et al.~ 1975), usually after breast feeding has been discontinued (Martin, 1978) the mode of infant feeding may not directly influence the colonisation of the infant's mouth and other factors such as diet may play a more significant role. Since the amount of saliva obtainable from the infants was limited, and antibody levels were in general rather low, it would be of considerable advantage if the sensitivity could be increased still more. Furthermore, the use of purified antigens rather than whole cells (Russell et al., 1980~ 1982; Lehner et al., 1985) should prove more specific than whole bacterial cells in the assay.
Acknowledgements The authors wish to thank Dr. R.G. Newcombe for advice regarding statistical analysis of the data. Supported by a grant from the Welsh Scheme for the Development of Health and Social Research.
References Aaltonen, A.S., J. Tenovuo, O.-P. Lehtonen, R. Saksala and O. Meurman, 1985, Arch. Oral Biol. 30, 331. Arnold, R.R., J. Mestecky and J.R. McGhee, 1976, Infect. Immun. 14, 355.
Bammann, L.L. and R.J. Gibbons, 1979, J. Clin. Microbiol. 10, 538. Berkowitz, R.J., H.V. Jordan and G. White, 1975, Arch. Oral Biol. 20, 171. Berkowitz, R.J., J. Turner and P. Green, 1980. Arch. Oral Biol. 25, 221. Bolton, R.W. and G . L Hlava, 1982, J. Dent. Res. 61, 1225. Catalanotto, F.A., I.L. Shklair and H.J. Keene. 1975, J. Am. Dent. Assoc. 91,606. Challacombe, S.J., 1978. Adv. Exp. Med. Biol. 107, 355. Challacombe, S.J., 1980, Arch. Oral Biol. 25,495. Challacombe, S.J. and T. Lehner, 1976, J. Dent. Res. 55, 139. Challacombe, S.J., B. Guggenheim and T. Lehner, 1973, Arch. Oral Biol. 18, 657. Challacombe, S.J., E.A. Bergmeier and A.S. Rees, 1984. Arch. Oral Biol. 29, 179. Czerkinsky, C., A.S. Rees, L A . Bergmeier and S.J. Challacombe, 1983, Clin. Exp. lmmunol. 53, 192. Eggert, F.M. and B.W. Gurner, 1984, Infect. Immun. 44, 660. Elder, B.L., D.K. Boraker and P.M. Fives-Taylor, 1982, J. Clin. Microbiol. 16, 141. Glass, R.I., A.-M. Svennerholm, B.J. Stoll, M.R Khan, K.M.B. Hossain, M.I. Huq and J. Holmgren, 1983, N. Engl. J. Med. 308, 1389. Huis in "t Veld, J., D. Bannet, W. Van Palenstein Helderman, P.S. Camargo and O. Backer-Dirks, 1978, Adv. Exp. Med. Biol. 107, 369. lvanyi, L. and T. Lehner, 1978, Arch. Oral Biol. 23, 851. Kennedy, A.E., I.L. Shklair, J.A. Hayashi and A.N. Bahn, 1968, Arch. Oral Biol. 13, 1275. KOhler, B., B.-M. Pettersson and D. Bratthall, 1981, Scand. J. Dent. Res. 89, 19. Lehner, T., J. Caldwell and S.J. Challacombe, 1977, Arch. Oral Biol. 27, 393. Lehner, T., J. Caldwell and A.S.M. Giassudin, 1985, Arch. Oral Biol. 30, 207. Lehtonen, O.-P.J., E.M. Grahn, T.H. Stahlberg and L.A. Laitinen, 1984, Infect. lmmun. 43, 308. McGhee, J.R. and S.M. Michalek, 1981, Ann. Rev. Microbiol. 35, 595. Magnusson, I. and T. Ericson, 1977, Caries Res. 11, 109. Martin, J., 1978, Infant Feeding 1975: Attitudes and Practice in England and Wales (H.M.S.O., London). Michael, J.G., R. Ringenback and S. Hottenstein, 1971, J. Infect. Dis. 124, 445. Ogra, S.S. and P.L. Ogra, 1978, J. Pediatr. 92, 546. Russell, M.W., L.A. Bergmeier, E.D. Zanders and T. Lehner, 1980, Infect. lmmun. 28, 486. Russell, R.R.B., D. Beighton and B. Cohen, 1982, Br. Dent. J. 152, 81. Tank, G. and C.A. Storvick, 1965, J. Am. Dent. Assoc. 70, 394. Totterdell, B.M., I.E. Chrystie and J.E. Banatvala, 1976, Arch. Dis. Child. 51,924. Voller, A. and D.E. Bidwell, 1975, Br. J. Exp. Pathol. 56, 338. Wade, W.G., D.R. Llewelyn, D.M. Walker and M.J. Aldred, 1983, J. Dent. Res. 62, 420. World Health Organisation, 1962, Technical Report Series No. 242, Standardisation of reporting on dental diseases and conditions, Report of an expert committee on dental health (W.H.O., Geneva).