Am J Otolaryngol 12:161-164.
1991
Immunoglobulin-Coated Bacteria in Effusions From Secretory and Chronic Suppurative Otitis Media LARS-ERIC
STENFORS, MD, PHD, AND SIMO RAIS~~NEN,MD, PhD
Samples of middle ear effusions from 10 children with secretory otitis media and from 10 children with chronic suppurative otitis media were subjected to qualitative and quantitative bacteriologic analysis. In addition, secretory immunoglobulin A (SIgA)- and IgGcoated bacteria were evaluated using the immunofluorescence technique. Secretory otitis media effusions harbored few, if any, immunoglobulin-coated bacteria, whereas chronic otitis media effusions as a rule had heavily IgG- and SlgA-coated bacteria. However, those chronic otitis media effusions that were culture positive for Pseudomonas aeruginosa had no immunoglobulin-coated bacteria. The effusions of very young children were completely devoid of SlgA-coated bacteria. This study demonstrates that, based on the immunoglobulin coating of bacteria obtained from the middle ear cleft, one can evaluate immunologic response during otitis media. AM J OTOLARYNGOL 12:161-164. Copyright 0 1991 by W.B. Saunders Company Key words: Immunoglobulins, opsonization, otitis media, fluorescence microscopy.
so-called secretory IgA (SIgA), is the first specific defensive barrier preventing bacteria from attaching to mucosal membranes.* There is abundant evidence that SIgA antibodies interfere with bacterial adherence to mucosal surfaces, thus limiting the colonization of these surfaces by pathogens,2-4 which is a prerequisite for invasive disease.536 In the presence of complement, IgG antibodies are reported to have both bactericidal and opsonic properties,7.8 whereas IgA antibodies have either a bactericidal or opsonic effect.g Immunoglobulin G constitutes approximately 80% of all serum immunoglobulins. For these immunoglobulins to act, the essential event is their attachment to the microorganisms, ie, the immunoglobulin coating of bacteria. This study was undertaken to ascertain the degree of the SIgA and IgG coating of bacteria obtained from the MEC during SOM and COM, respectively. Particular effort was made to get young individuals in both groups, as concentrations of serum immunoglobulins vary considerably according to age.” Our findings are discussed in relation to recent reports on the presence of immunoglobulins in middle ear effusions (MEEs).“,~’
It is well established that bacteria play a crucial role in the majority of middle ear infections. Microorganisms located in the external auditory meatus can gain access to the middle ear cavity (MEC) by way of the tympanic membrane, that is, through a pre-existing perforation or a ventilation tube. Another possible approach is via the eustachian tube, through which middle ear pathogens can ascend from the nasopharynx. The latter mode is certainly operative in cases of secretory otitis media (SOM) with an intact drum, whereas either route might be implicated in the etiology of chronic suppurative otitis media (COM) with a discharging ear. When microorganisms enter an otherwise sterile MEC, the host defense may involve the mucociliary system, the immunologic system, and the phagocytic system. It has never been established which of these systems is most important to middle ear defense. However, in recent years, much interest has focused on the role of the local mucosal immune system. Locally synthesized immunoglobulin A (IgA), Received March 10, 1991. from the Department of Otolaryngology, University of TromsB, TromsB, Norway; the Department of Otolaryngology, Central Hospital of Keski-Pohjanmaa, Kokkola. Finland; and the Clinical Laboratory, Central Hospital of Keski-Pohjanmaa, Kokkola, Finland. Accepted for publication May 2, 1991. Address correspondence and reprint requests to Lars-Eric Stenfors, MD, Department of Otolaryngology, Central Hospital of Keski-Pohjanmaa, SF-6700 Kokkola, Finland. Copyright 0 1991 by W.B. Saunders Company 0196-0709/91/1203-0009$5.00/O
MATERIALS
AND METHODS
This study involved 20 young patients, 10 with SOM and 10 with COM. The patients’ ages and sex are shown in Tables 1 and 2. All SOM patients had intact drums. Middle ear effusions were obtained while patients were
161
162
IMMUNOGLOBULIN-COATED
TABLE 1.
Characteristics
in the Secretory
of Individuals
Otitis Media Group INTENSITY OF BACTERIA INTHEMEE
BACTERIAL FINDINGS PATIENT No.
AGE
1 2
1 yr 3 mo lyr4mo
3
2 yr 7 mo
SEX ;
Streptococcus S pneumoniae S pneumoniae Haemophilus BranhameIIa B H H B H H
catarrhalis influenzae influenzae catarrhalis influenzae influenzae
30 1 10 50 50 8
H H H S
influenzae influenzae influenzae pneumoniae
1 10 40 1
4
3 yr 1 mo
(dx) (sin) M (dxl
5
3 yr 5 mo
(sin) F (dxl
6 7 8
4 yr 1 mo 4yrlmo 5 yr 3 mo
M (sin1 F M (dxl
9
8yr7mo
M (dx) (sin] F
(sin)
10 Abbreviations:
9 yr 2 mo dx, dexter;
Count(X 104)/ mL MEE
Species pneumoniae
influenzae catarrhalis
BACTERIA
SIgA-Coated
IgG-Coated
1 50
_ _
_ _
6 4 6
+ _ _ _
++ _ _
_ _
+ + _
_ _ _ _ _ _ -
+ + _ _ _ _ _
sin, sinister.
under general anesthesia for the insertion of a tympanostomy tube and/or for adenoidectomy. Prior to sampling, the external auditory meatus was cleansed of wax and detritus, and washed twice with a 70% alcohol solution. A myringotomy was made in the anterior/ inferior quadrant and a cannula connected to a syringe was inserted through the drum perforation. Mucoid (glue-like, extremely hydrophilic) MEE was aspirated from the MEC in every case. Only SOM patients with culture-positive MEE were included in the study. From the patients with COM, indicated by drainage from the ear and a drum perforation, MEE was aspirated directly from the MEC via the perforation into a sterile syringe. Specimens were immediately transferred to the clinical laboratory for further processing. At the laboratory, the following procedures were performed: 1. Part of the MEE was spread over blood- and chocolate-agar plates for standard aerobic culturing. 2. The remaining MEE was spread evenly over clean glass slides, allowed to dry, and fixed with 70% alcohol solution. One sample was stained with acridine orange and examined under a fluorescence microscope to facilitate quantification of the bacteria according to a reTABLE 2.
Characteristics
cently described method.13 The remaining slides were stored at -20°C for later immunofluorescence processing. 3. After thawing, two hydrophobic rings were drawn on each dried glass slide. Inside each ring, two drops of fluorescein-conjugated goat anti-human SIgA (alpha chain specific; Cappel, Organon Teknika Corporation #13Ol-0011, West Chester, PA) and goat anti-human IgG (heavy and light chain specific; Cappel, Organon Teknika Corporation #1201-0081) were added, respectively. Counterstaining was performed with 0.25% Evans blue. All samples were processed simultaneously, using the same reagent and its several dilutions. The specimens were examined under a Leitz fluorescence microscope. Prior to the immunofluorescence studies, the antisera were titrated according to principles promulgated by the US Department of Health, Education, and Welfare.14 The findings were graded as follows: -, bacteria stained red, signifying no antibody coating (Fig 1); + , the core of the bacteria stained red, but the periphery stained green; + +, the periphery of the bacteria stained intensely green (Fig 2); and + + + , the whole bacteria stained bright green.
of Individuals
in the Chronic Otitis Media INTENSITY OF BACTERIA IN THE
BACTERIAL FINDINGS PATIENT No.
AGE
SEX
11 12
1 yr 1 mo
1 yr 1 mo
F F (dx1
13
1 yr 1 mo
Mcsinl
14 15 16 17 18 19 20
lyr8mo 6yr8mo 7yrZmo 9yr6mo 9yr7mo 10 yr 3 mo 13 yr 1 mo
M M F
M M F F
Abbreviations: dx, dexter: sin, sinister.
Species Haemophilus influenzae Staphylococcus aureus S aureus S aureus Streptococcus pneumoniae S aureus Pseudomonas aeruginosa S aureus P aeruginosa S aureus S aureus, diphtheroides S aureus
Count (X 104)/ mL MEE
SIgA-Coated
MEE IgG-Coated
80,000 100,000 100,000 800
_ _ _ _
++ ++ ++ ++
800 1,000 500 1,000 100,000 100,000 40,000
++ _ + _ ++ ++ ++
++ _ + _ ++ ++ ++
STENFORS
163
AND RAISliNEN
Figure 1. Photograph of an MEE sample obtained from patient no. 17, who had a growth of Pseudomonos oeruginosa. The bacteria stained bright red, indicating no SIgA coating (arrows). (Direct FA staining with FITC-labeled goat antihuman SIgA serum. counterstained with 0.25% Evans blue; magnification x 3.500.) (Created from original color figure.)
Table 2 shows that four of the 11 (36%) MEE samples from COM patients harbored bacteria with a rich peripheral SIgA coating of bacteria. In six of these (55%), no SIgA-coated bacteria were found. Four of these six samples were from the youngest patients in this group; all were less than 1 year 8 months old. Eight (73%) of the MEE samples had richly IgG-coated bacteria ( + +) (Fig 2). In addition, four of the MEE samples (36%) harbored bacteria that were intensely coated with SIgA and IgG antibodies simultaneously. The predominant bacterial species was Staphylococcus aureus, followed by Pseudomonas aeruginosa. It is notable that the two samples harboring P aeruginosa showed no antibody-coated bacteria (Fig 1). The bacterial counts ranged from 5 x lo7 to 10' bacteria/ml MEE. DISCUSSION
RESULTS
Table 1 shows that in the majority of the culture-positive SOM samples, no immunoglobulincoated bacteria were found. Only one MEE sample showed any SIgA-coated bacteria (Streptococcus pneumoniae); the fluorescence intensity was rather weak (+]. The periphery of these bacteria showed intense fluorescence, however, when stained with FITC-labeled anti-IgG antibodies. Thirty-three percent (five of 15)of the MEE samples harbored bacteria coated to some extent with IgG. The predominant bacteria1 species was Haemophilus influenzae (eight of 151, followed by S pneumoniae (four of 15) and Branhamella catarrhalis (three of 19). The bacterial counts ranged from 5 x 10' to lo5 bacteria/ml. It must be emphasized that SOM patients with culture-negative effusions were excluded from the study.
Figure 2. Photograph of an MEE sample obtained from pa(+ +) tient no. 18 showing Staphylococcus aureus heavily coated with IgG (arrows). The periphery of the bacteria stained intensely green. (Direct FA staining with FITC-labeled goat anti-human IgG serum. counterstained with 0.25% Evans blue; magnification x 3,500.) (Created from original color figure.)
It is we11 established that the mucosa of the MEC is immunocompetent and can mount a local immune response similar to those observed in other mucosal surfaces of the upper respiratory tract when infected by otitis media.11~‘2 This study demonstrates a distinct difference in SIgA- and IgG-coated bacteria between SOM and COM. Secretory IgA-coated bacteria were scarce in the extremely mucoid and hydrophilic MEE. Moreover, IgG-coated bacteria, if present, were very sparse. It should be emphasized that SIgA must be considered to be produced locally by plasma cells in the middle ear mucosa,12 whereas IgG gains access to the MEC by passive leakage through the epithelium.15 In COM, intense SIgA and IgG coating of MEC pathogens was common. With regard to SIgA coating, two remarkable observations were noted. First, young children harbored no SIgA-coated bacteria in the MEC. Second, when P aeruginosa was the causative agent of the middle ear infection, no bacterial coating, either SIgA or IgG, was seen. In contrast, when S aureus was the causative agent, immunoglobulin-coated bacteria were abundant, with the exception of SIgA-coated bacteria in the youngest children. Secretory IgA is known to be a significant agent in preventing bacterial adherence to mucosal cells, an important precondition of bacterial colonization and invasive disease.2,3.6 One immediate cause of the development of otitis media might be inadequate SIgA coating. This important finding must be considered when trying to prevent middle ear infections by vaccination.‘6*‘7 The main function of IgG-class immunoglobulins is to promote lysis of the bacteria or to opsonize the bacteria by activating the complement
164 system via the classic pathway.7-g Tables 1 and 2 show that the bacterial count in the MEE of SOM is roughly l/10,000 of that in COM. In our materials, middle ear pathogens were found in mucoid effusion in 22% of SOM patients.” However, in a recent study,lg an additional 19% of the cases exhibited dormant S pneumoniae and II influenzae in mucoid MEE. These findings could be attributed to an inadequate coating of bacteria with IgG, thereby hindering the initiation of normal defenses, that is, phagocytosis of the bacteria entering the otherwise sterile MEC. This hypothesis is consistent with the clinical persistence of SOM. The immunofluorescence technique used in this study offers evident advantages to the researcher seeking to elucidate the role of immunoglobulins in the MEEs associated with various middle ear infections. In measuring the immunoglobulins attached to middle ear pathogens, one can be sure that the antibodies originate from the MEC and not from the blood and/or serum that is unavoidably added to the aspirated middle ear material following a myringotomy. However, several questions remain, one of the most important being the amount of specific immunoglobulin necessary to stimulate an adequate physiologic response. References 1. Russell MW, Mestecky J: Induction of the mucosal immune response. Rev Infect Dis 1988; 10:440-446 (suppl 2) 2. Tomasi TB, Gray HM: Structure and function of immunoglobulin A. Prog Allergy 1972; 16:18-213 3. Williams RC, Gibbons RJ: Inhibition of bacterial adherence by secretory immunoglobulin A: A mechanism of antigen disposal. Science 1972; 177:697-699 4. Fubara ES, Freter R: Protection against enteric bacterial infection by secretory IgA antibodies. J Immunol 1973; 3:395403
IMMLJNOGLOBULIN-COATED BACTERIA 5. Abraham SN, Beachey EH: Host defence against adhesion of bacteria to mucosal surfaces, in Gallin JI, Fauci AS (eds): Advances in Host Defence Mechanisms, ~014. New York, NY, Raven, 1985, pp 63-88 6. Stenfors L-E, Rlislnen S: In vivo attachment of betahaemolytic streptococci to tonsillar epithelial cells in health and disease. Acta Otolaryngol (Stockh) 1991; 111:562-568 7. Johnston RB, Jr, Anderson P, Rosen FS, et al: Characterization of human antibody to polyribophosphate, the capsular antigen of Haemophilus influenzae, type b. Clin Immunol Immunopathol 1973: 1234-240 8. Schreiber JR, Barrus V, Cates KL, et al: Functional characterization of human IgG, IgM and IgA antibody directed to capsule of Haemophilus influenzae type b. J Infect Dis 1986; 153:8-16 9. Amir J, Scott MG, Nahm MH, et al: Bactericidal and opsonic activity of IgGl and IgG2 anticapsular antibodies to Haemophilus influenzae type b. J infect Dis 1990; 162:163-171 10. Johansson SGO, Berg T: Immunoglobulin levels in healthy children. Acta Paediatr Stand 1967; 56:572-579 11. Ogra PL, Bernstein JM, Yurchar AM, et al: Characteristics of secretory immune system in human middle ear: Implications in otitis media. J Immunol 1974; 112,4R3-495 _- _ - - 12. Mogi G, Honjo S, Maeda S, et al: Secretory immunoglobulin A (SIgA) in middle ear effusions. Ann Otol Rhino1 Laryngo1 1974; 83:92-101 13. Stenfors L-E, Raisanen S: Quantification of bacteria in middle ear effusions. Acta Otolaryngol (Stockh) 1988; 106:435440 14. Laboratory outline for training course in principles and bacterial applications of fluorescent antibody techniques. US Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control, Atlanta, GA, 1972 15. Brandtzaeg P: Immunobarriers of the mucosa of the upper respiratory and digestive pathways. Acta Otolaryngol [Stockh) 1988; 106:172-180 16. Austrian R: Prevention of pneumococcal infection by immunization with capsular polysaccharides of Streptococcus pneumoniae: Current status of polyvalent vaccines. J Infect Dis 1977; 136:38-42 (suppl) 17. Rosen C, Christensen P, Henrichsen J, et al: Beneficial effect of pneumococcal vaccination on otitis media in children over two years old. Int J Pediatr Otorhinolaryngol 1984; 7:239246 18. Stenfors L-E, Rlisanen S: Quantitative analysis of the bacterial findings in otitis media. J Laryngol Otol 1990; 104:749-757 19. Stenfors L-E, Raisanen S: Occurrence of Streptococcus pneumoniae and Haemophilus influenzae in otitis media with effusion. Clin Otolaryngol (in press)
1991
Immunoglobulin-Coated Bacteria in Effusions From Secretory and Chronic Suppurative Otitis Media LARS-ERIC
STENFORS, MD, PHD, AND SIMO RAIS~~NEN,MD, PhD
Samples of middle ear effusions from 10 children with secretory otitis media and from 10 children with chronic suppurative otitis media were subjected to qualitative and quantitative bacteriologic analysis. In addition, secretory immunoglobulin A (SIgA)- and IgGcoated bacteria were evaluated using the immunofluorescence technique. Secretory otitis media effusions harbored few, if any, immunoglobulin-coated bacteria, whereas chronic otitis media effusions as a rule had heavily IgG- and SlgA-coated bacteria. However, those chronic otitis media effusions that were culture positive for Pseudomonas aeruginosa had no immunoglobulin-coated bacteria. The effusions of very young children were completely devoid of SlgA-coated bacteria. This study demonstrates that, based on the immunoglobulin coating of bacteria obtained from the middle ear cleft, one can evaluate immunologic response during otitis media. AM J OTOLARYNGOL 12:161-164. Copyright 0 1991 by W.B. Saunders Company Key words: Immunoglobulins, opsonization, otitis media, fluorescence microscopy.
so-called secretory IgA (SIgA), is the first specific defensive barrier preventing bacteria from attaching to mucosal membranes.* There is abundant evidence that SIgA antibodies interfere with bacterial adherence to mucosal surfaces, thus limiting the colonization of these surfaces by pathogens,2-4 which is a prerequisite for invasive disease.536 In the presence of complement, IgG antibodies are reported to have both bactericidal and opsonic properties,7.8 whereas IgA antibodies have either a bactericidal or opsonic effect.g Immunoglobulin G constitutes approximately 80% of all serum immunoglobulins. For these immunoglobulins to act, the essential event is their attachment to the microorganisms, ie, the immunoglobulin coating of bacteria. This study was undertaken to ascertain the degree of the SIgA and IgG coating of bacteria obtained from the MEC during SOM and COM, respectively. Particular effort was made to get young individuals in both groups, as concentrations of serum immunoglobulins vary considerably according to age.” Our findings are discussed in relation to recent reports on the presence of immunoglobulins in middle ear effusions (MEEs).“,~’
It is well established that bacteria play a crucial role in the majority of middle ear infections. Microorganisms located in the external auditory meatus can gain access to the middle ear cavity (MEC) by way of the tympanic membrane, that is, through a pre-existing perforation or a ventilation tube. Another possible approach is via the eustachian tube, through which middle ear pathogens can ascend from the nasopharynx. The latter mode is certainly operative in cases of secretory otitis media (SOM) with an intact drum, whereas either route might be implicated in the etiology of chronic suppurative otitis media (COM) with a discharging ear. When microorganisms enter an otherwise sterile MEC, the host defense may involve the mucociliary system, the immunologic system, and the phagocytic system. It has never been established which of these systems is most important to middle ear defense. However, in recent years, much interest has focused on the role of the local mucosal immune system. Locally synthesized immunoglobulin A (IgA), Received March 10, 1991. from the Department of Otolaryngology, University of TromsB, TromsB, Norway; the Department of Otolaryngology, Central Hospital of Keski-Pohjanmaa, Kokkola. Finland; and the Clinical Laboratory, Central Hospital of Keski-Pohjanmaa, Kokkola, Finland. Accepted for publication May 2, 1991. Address correspondence and reprint requests to Lars-Eric Stenfors, MD, Department of Otolaryngology, Central Hospital of Keski-Pohjanmaa, SF-6700 Kokkola, Finland. Copyright 0 1991 by W.B. Saunders Company 0196-0709/91/1203-0009$5.00/O
MATERIALS
AND METHODS
This study involved 20 young patients, 10 with SOM and 10 with COM. The patients’ ages and sex are shown in Tables 1 and 2. All SOM patients had intact drums. Middle ear effusions were obtained while patients were
161
162
IMMUNOGLOBULIN-COATED
TABLE 1.
Characteristics
in the Secretory
of Individuals
Otitis Media Group INTENSITY OF BACTERIA INTHEMEE
BACTERIAL FINDINGS PATIENT No.
AGE
1 2
1 yr 3 mo lyr4mo
3
2 yr 7 mo
SEX ;
Streptococcus S pneumoniae S pneumoniae Haemophilus BranhameIIa B H H B H H
catarrhalis influenzae influenzae catarrhalis influenzae influenzae
30 1 10 50 50 8
H H H S
influenzae influenzae influenzae pneumoniae
1 10 40 1
4
3 yr 1 mo
(dx) (sin) M (dxl
5
3 yr 5 mo
(sin) F (dxl
6 7 8
4 yr 1 mo 4yrlmo 5 yr 3 mo
M (sin1 F M (dxl
9
8yr7mo
M (dx) (sin] F
(sin)
10 Abbreviations:
9 yr 2 mo dx, dexter;
Count(X 104)/ mL MEE
Species pneumoniae
influenzae catarrhalis
BACTERIA
SIgA-Coated
IgG-Coated
1 50
_ _
_ _
6 4 6
+ _ _ _
++ _ _
_ _
+ + _
_ _ _ _ _ _ -
+ + _ _ _ _ _
sin, sinister.
under general anesthesia for the insertion of a tympanostomy tube and/or for adenoidectomy. Prior to sampling, the external auditory meatus was cleansed of wax and detritus, and washed twice with a 70% alcohol solution. A myringotomy was made in the anterior/ inferior quadrant and a cannula connected to a syringe was inserted through the drum perforation. Mucoid (glue-like, extremely hydrophilic) MEE was aspirated from the MEC in every case. Only SOM patients with culture-positive MEE were included in the study. From the patients with COM, indicated by drainage from the ear and a drum perforation, MEE was aspirated directly from the MEC via the perforation into a sterile syringe. Specimens were immediately transferred to the clinical laboratory for further processing. At the laboratory, the following procedures were performed: 1. Part of the MEE was spread over blood- and chocolate-agar plates for standard aerobic culturing. 2. The remaining MEE was spread evenly over clean glass slides, allowed to dry, and fixed with 70% alcohol solution. One sample was stained with acridine orange and examined under a fluorescence microscope to facilitate quantification of the bacteria according to a reTABLE 2.
Characteristics
cently described method.13 The remaining slides were stored at -20°C for later immunofluorescence processing. 3. After thawing, two hydrophobic rings were drawn on each dried glass slide. Inside each ring, two drops of fluorescein-conjugated goat anti-human SIgA (alpha chain specific; Cappel, Organon Teknika Corporation #13Ol-0011, West Chester, PA) and goat anti-human IgG (heavy and light chain specific; Cappel, Organon Teknika Corporation #1201-0081) were added, respectively. Counterstaining was performed with 0.25% Evans blue. All samples were processed simultaneously, using the same reagent and its several dilutions. The specimens were examined under a Leitz fluorescence microscope. Prior to the immunofluorescence studies, the antisera were titrated according to principles promulgated by the US Department of Health, Education, and Welfare.14 The findings were graded as follows: -, bacteria stained red, signifying no antibody coating (Fig 1); + , the core of the bacteria stained red, but the periphery stained green; + +, the periphery of the bacteria stained intensely green (Fig 2); and + + + , the whole bacteria stained bright green.
of Individuals
in the Chronic Otitis Media INTENSITY OF BACTERIA IN THE
BACTERIAL FINDINGS PATIENT No.
AGE
SEX
11 12
1 yr 1 mo
1 yr 1 mo
F F (dx1
13
1 yr 1 mo
Mcsinl
14 15 16 17 18 19 20
lyr8mo 6yr8mo 7yrZmo 9yr6mo 9yr7mo 10 yr 3 mo 13 yr 1 mo
M M F
M M F F
Abbreviations: dx, dexter: sin, sinister.
Species Haemophilus influenzae Staphylococcus aureus S aureus S aureus Streptococcus pneumoniae S aureus Pseudomonas aeruginosa S aureus P aeruginosa S aureus S aureus, diphtheroides S aureus
Count (X 104)/ mL MEE
SIgA-Coated
MEE IgG-Coated
80,000 100,000 100,000 800
_ _ _ _
++ ++ ++ ++
800 1,000 500 1,000 100,000 100,000 40,000
++ _ + _ ++ ++ ++
++ _ + _ ++ ++ ++
STENFORS
163
AND RAISliNEN
Figure 1. Photograph of an MEE sample obtained from patient no. 17, who had a growth of Pseudomonos oeruginosa. The bacteria stained bright red, indicating no SIgA coating (arrows). (Direct FA staining with FITC-labeled goat antihuman SIgA serum. counterstained with 0.25% Evans blue; magnification x 3.500.) (Created from original color figure.)
Table 2 shows that four of the 11 (36%) MEE samples from COM patients harbored bacteria with a rich peripheral SIgA coating of bacteria. In six of these (55%), no SIgA-coated bacteria were found. Four of these six samples were from the youngest patients in this group; all were less than 1 year 8 months old. Eight (73%) of the MEE samples had richly IgG-coated bacteria ( + +) (Fig 2). In addition, four of the MEE samples (36%) harbored bacteria that were intensely coated with SIgA and IgG antibodies simultaneously. The predominant bacterial species was Staphylococcus aureus, followed by Pseudomonas aeruginosa. It is notable that the two samples harboring P aeruginosa showed no antibody-coated bacteria (Fig 1). The bacterial counts ranged from 5 x lo7 to 10' bacteria/ml MEE. DISCUSSION
RESULTS
Table 1 shows that in the majority of the culture-positive SOM samples, no immunoglobulincoated bacteria were found. Only one MEE sample showed any SIgA-coated bacteria (Streptococcus pneumoniae); the fluorescence intensity was rather weak (+]. The periphery of these bacteria showed intense fluorescence, however, when stained with FITC-labeled anti-IgG antibodies. Thirty-three percent (five of 15)of the MEE samples harbored bacteria coated to some extent with IgG. The predominant bacteria1 species was Haemophilus influenzae (eight of 151, followed by S pneumoniae (four of 15) and Branhamella catarrhalis (three of 19). The bacterial counts ranged from 5 x 10' to lo5 bacteria/ml. It must be emphasized that SOM patients with culture-negative effusions were excluded from the study.
Figure 2. Photograph of an MEE sample obtained from pa(+ +) tient no. 18 showing Staphylococcus aureus heavily coated with IgG (arrows). The periphery of the bacteria stained intensely green. (Direct FA staining with FITC-labeled goat anti-human IgG serum. counterstained with 0.25% Evans blue; magnification x 3,500.) (Created from original color figure.)
It is we11 established that the mucosa of the MEC is immunocompetent and can mount a local immune response similar to those observed in other mucosal surfaces of the upper respiratory tract when infected by otitis media.11~‘2 This study demonstrates a distinct difference in SIgA- and IgG-coated bacteria between SOM and COM. Secretory IgA-coated bacteria were scarce in the extremely mucoid and hydrophilic MEE. Moreover, IgG-coated bacteria, if present, were very sparse. It should be emphasized that SIgA must be considered to be produced locally by plasma cells in the middle ear mucosa,12 whereas IgG gains access to the MEC by passive leakage through the epithelium.15 In COM, intense SIgA and IgG coating of MEC pathogens was common. With regard to SIgA coating, two remarkable observations were noted. First, young children harbored no SIgA-coated bacteria in the MEC. Second, when P aeruginosa was the causative agent of the middle ear infection, no bacterial coating, either SIgA or IgG, was seen. In contrast, when S aureus was the causative agent, immunoglobulin-coated bacteria were abundant, with the exception of SIgA-coated bacteria in the youngest children. Secretory IgA is known to be a significant agent in preventing bacterial adherence to mucosal cells, an important precondition of bacterial colonization and invasive disease.2,3.6 One immediate cause of the development of otitis media might be inadequate SIgA coating. This important finding must be considered when trying to prevent middle ear infections by vaccination.‘6*‘7 The main function of IgG-class immunoglobulins is to promote lysis of the bacteria or to opsonize the bacteria by activating the complement
164 system via the classic pathway.7-g Tables 1 and 2 show that the bacterial count in the MEE of SOM is roughly l/10,000 of that in COM. In our materials, middle ear pathogens were found in mucoid effusion in 22% of SOM patients.” However, in a recent study,lg an additional 19% of the cases exhibited dormant S pneumoniae and II influenzae in mucoid MEE. These findings could be attributed to an inadequate coating of bacteria with IgG, thereby hindering the initiation of normal defenses, that is, phagocytosis of the bacteria entering the otherwise sterile MEC. This hypothesis is consistent with the clinical persistence of SOM. The immunofluorescence technique used in this study offers evident advantages to the researcher seeking to elucidate the role of immunoglobulins in the MEEs associated with various middle ear infections. In measuring the immunoglobulins attached to middle ear pathogens, one can be sure that the antibodies originate from the MEC and not from the blood and/or serum that is unavoidably added to the aspirated middle ear material following a myringotomy. However, several questions remain, one of the most important being the amount of specific immunoglobulin necessary to stimulate an adequate physiologic response. References 1. Russell MW, Mestecky J: Induction of the mucosal immune response. Rev Infect Dis 1988; 10:440-446 (suppl 2) 2. Tomasi TB, Gray HM: Structure and function of immunoglobulin A. Prog Allergy 1972; 16:18-213 3. Williams RC, Gibbons RJ: Inhibition of bacterial adherence by secretory immunoglobulin A: A mechanism of antigen disposal. Science 1972; 177:697-699 4. Fubara ES, Freter R: Protection against enteric bacterial infection by secretory IgA antibodies. J Immunol 1973; 3:395403
IMMLJNOGLOBULIN-COATED BACTERIA 5. Abraham SN, Beachey EH: Host defence against adhesion of bacteria to mucosal surfaces, in Gallin JI, Fauci AS (eds): Advances in Host Defence Mechanisms, ~014. New York, NY, Raven, 1985, pp 63-88 6. Stenfors L-E, Rlislnen S: In vivo attachment of betahaemolytic streptococci to tonsillar epithelial cells in health and disease. Acta Otolaryngol (Stockh) 1991; 111:562-568 7. Johnston RB, Jr, Anderson P, Rosen FS, et al: Characterization of human antibody to polyribophosphate, the capsular antigen of Haemophilus influenzae, type b. Clin Immunol Immunopathol 1973: 1234-240 8. Schreiber JR, Barrus V, Cates KL, et al: Functional characterization of human IgG, IgM and IgA antibody directed to capsule of Haemophilus influenzae type b. J Infect Dis 1986; 153:8-16 9. Amir J, Scott MG, Nahm MH, et al: Bactericidal and opsonic activity of IgGl and IgG2 anticapsular antibodies to Haemophilus influenzae type b. J infect Dis 1990; 162:163-171 10. Johansson SGO, Berg T: Immunoglobulin levels in healthy children. Acta Paediatr Stand 1967; 56:572-579 11. Ogra PL, Bernstein JM, Yurchar AM, et al: Characteristics of secretory immune system in human middle ear: Implications in otitis media. J Immunol 1974; 112,4R3-495 _- _ - - 12. Mogi G, Honjo S, Maeda S, et al: Secretory immunoglobulin A (SIgA) in middle ear effusions. Ann Otol Rhino1 Laryngo1 1974; 83:92-101 13. Stenfors L-E, Raisanen S: Quantification of bacteria in middle ear effusions. Acta Otolaryngol (Stockh) 1988; 106:435440 14. Laboratory outline for training course in principles and bacterial applications of fluorescent antibody techniques. US Department of Health, Education, and Welfare, Public Health Service, Center for Disease Control, Atlanta, GA, 1972 15. Brandtzaeg P: Immunobarriers of the mucosa of the upper respiratory and digestive pathways. Acta Otolaryngol [Stockh) 1988; 106:172-180 16. Austrian R: Prevention of pneumococcal infection by immunization with capsular polysaccharides of Streptococcus pneumoniae: Current status of polyvalent vaccines. J Infect Dis 1977; 136:38-42 (suppl) 17. Rosen C, Christensen P, Henrichsen J, et al: Beneficial effect of pneumococcal vaccination on otitis media in children over two years old. Int J Pediatr Otorhinolaryngol 1984; 7:239246 18. Stenfors L-E, Rlisanen S: Quantitative analysis of the bacterial findings in otitis media. J Laryngol Otol 1990; 104:749-757 19. Stenfors L-E, Raisanen S: Occurrence of Streptococcus pneumoniae and Haemophilus influenzae in otitis media with effusion. Clin Otolaryngol (in press)
