THE JOURNAL OF INFECTIOUS DISEASES • VOL. 133, NO.2· FEBRUARY 1976 © 1976 by the University of Chicago. All rights reserved.
Immunochemistry of Otitis Media with Effusion From the Department of Otolaryngology, Yamaguchi University School of Medicine, Ube, Yamaguchi-ken, Japan
Goro Mogi, Shoichi Maeda, Toyoharu Yoshida, and Noritake Watanabe
Within recent years, striking biochemical and immunochemical analytic techniques have been developed and have opened a new approach to the study of otitis media with effusion. This study was designed to examine immunoglobulin components of middle ear effusions, particularly secretory immunoglobulin A, for clarification of the nature of fluid in the middle ear cavity and for investigation of the existence of a local immune system in the middle ear.
when the middle ear effusion was aspirated. These patients were between six and 78 years of age; 50% were> 60 years old, 35% were between 17 and 59 years old, and 15 % were < 16 years old. Middle ear effusions were classified as serous type or mucoid type by gross appearance; clear fluid with a low viscosity similar to that of serum was referred to as serous, whereas viscous fluid was referred to as mucoid type. Our mucoid type involved the mucopurulent type classified by Senturia [1]. This type of effusion was found in seven cases, which were subjected to only qualitative analysis of secretory IgA. The aspirates were also broken down by origin from an acute case or from a chronic case. Acute cases had short duration of illness (less than three weeks) and transient existence of the effusion, whereas chronic cases had long duration of illness (more than three weeks) and recurrent existence of the effusion. The amount of ear aspirate varied among patients, but it was usually between 0.1 and 0.6 ml, All specimens of middle ear effusion and serum were kept in a deep freezer (-20 C) until use. Once the specimen had been melted to use for one of the procedures in the study, this sample was not used in any other procedure except affinity chromatography. Therefore, each procedure was usually performed upon one aspirate.
Materials and Methods
Middle ear effusions were collected by aspiration through the eardrums of more than 400 patients diagnosed as having otitis media with effusion. In all cases, acute signs of otitis media, such as earache and hyperemia of the eardrum, were absent
Received for publication March 12, 1975, and in revised form October 1, 1975. This work was supported by grant no. 848193 from the Ministry of Education, Japan. We wish to thank Drs. T. Nakashima, M. Ogata, Y. Tawara, and Y. Onaka for providing specimens. Please address requests for reprints to Dr. Goro Mogi, Department of Otolaryngology, Yamaguchi University School of Medicine, Ube, Yamaguchi-ken, Japan.
126
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
For investigation of the nature and origin of middle ear effusions, immunochemical studies were performed on more than 400 patients diagnosed as having otitis media with effusion. Although results of cellulose acetate and disc electrophoretic analyses and quantitation of IgG, IgA, and IgM suggested that proteins found in the effusions were derived for the most part from the serum, quantitative analysis of secretory IgA revealed the existence of appreciable amounts of secretory IgA in both serous and mucoid effusions. The antigenicity and subunit structure of the secretory IgA isolated from middle ear effusions were identical or very similar to those of secretory IgA obtained from other external secretions. Radioactive single radial diffusion analysis of IgE showed that the mean concentrations of IgE in effusions and sera were within normal ranges. Findings of this study suggest that the middle ear maintains the local immunologic defense system, that the middle ear effusion is at least partially an external secretion, and that IgE in middle ear effusions obtained from old patients, rather than being a local product, may be derived from the serum.
Otitis Media with Effusion
methods described by Tomasi et al. [4], Mestecky et al. [5], and Kobayashi [6]. Antisera to secretory IgA, free secretory component, and a-chain. Production of antiserum to human secretory IgA was induced by immunization of rabbits with isolated secretory IgA. The rabbit antiserum to secretory IgA was absorbed with cord serum to make mono specific antiserum. Antiserum to secretory component was made by absorption of the antiserum to secretory IgA with pooled normal human sera and by immunization of rabbits with free secretory component. Rabbit antiserum to a-chain was obtained from Behringwerke, Marburg-Lahn, Germany. Preparation of 7S IgA. The 7S IgA (serum IgA) was purified from the serum of a patient with IgA myeloma. The crude y-globulins were precipitated with (NH4) 2S04 at 40% final saturation. The precipitate was dissolved again and dialyzed against 0.01 M phosphate buffer at pH 7.5. The crude y-globulins were subjected to ionexchange chromatography on DEAE cellulose. Eluates in 0.1 M phosphate buffer (pH 6.2) were concentrated, dialyzed against saline buffered with Tris-HCI (pH 7.4), and filtered through Seph arose 6B gel packed in a 1.5- X 100-cm column. The main fraction, appearing as the second peak, was rich in 7S IgA. Isolation of secretory IgA from middle ear effusions. Affinity chromatography was used to isolate secretory IgA from the pooled middle ear effusion. This technique was performed as described by Wofsy and Burr [7]. Sepharose 6B gel coupled to antibodies to secretory component was packed to a height of 8 em in a column (1.5X 30-cm), and 10 ml of the pooled middle ear effusion was applied. After nonspecific proteins were washed out, eluates were adjusted to neutral pH with 0.17 M glycine-Hel buffer (pH 2.3), concentrated, and passed through Sepharose 6B gel. The fraction containing secretory IgA was collected and concentrated. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis. For investigation of the constituents of the secretory IgA obtained from middle ear effusions and other external secretions, and for determination of the molecular weights of these polypeptide chains, sodium dodecyl sulfate (SDS)polyacrylamide gel electrophoresis was performed according to the method of Weber and Osborn
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
A blood sample for serum was taken from each patient at the same time that the effusion was aspirated. Human colostrum was provided by the Department of Gynecology and Obstetrics, Yamaguchi University School of Medicine. Parotid secretions were collected from normal adult humans by insertion of a capillary tube into the Stensen's ducts. Nasal secretions were collected from patients with chronic rhinitis or chronic paranasal sinus infection. These secretions were the source of secretory IgA and free secretory component. Determination of total protein concentration. The total concentration of protein was measured by a Hitachi serum protein meter (refractometer) (Tsukasa Kogyu, Nikko, Ibaragi, Japan). Quantitative analysis of protein fractions. Microzone electrophoresis on cellulose acetate (Separaxw; Joko Sangyo, Co., Ltd., Bunkyo-ku, Tokyo, Japan) and densitometry were used for the quantitative analysis. Cellulose acetate electrophoresis was performed as outlined by Kohn [2] with use of 0.07 M Veronal buffer at pH 8.6. The quantitative determination of the separate components, which were stained with Ponceau's 3R (Matheson Co., Elk Grove Village, Ill.), was performed with use of an automatic recording and integrating scanner (Kayagaki Instruments, Bunkyo-ku, Tokyo, Japan). Disc electrophoresis. Disc electrophoresis was performed according to the method of Davis [3] in 7.5% polyacrylamide gel equilibrated with Tris-HCI buffer (pH 8.9) with a constant current of 2 rnA per tube for 90 min. Samples of 3 or 4 III of middle ear effusion or sera were used in each tube. After the completion of electrophoresis, the gels were removed from the glass tubes, placed into a staining solution of 1% amido Schwarz® dye (E. Merck, Darmstadt, B.R.-Germany) in 7% acetic acid, and washed with 7 % acetic acid to remove excess dye. Quantitative analysis of IgG, IgA, and IgM. Levels of IgG, IgA, and IgM in both middle ear effusion and sera were determined with Hyland immunoplates (Hyland Laboratories, Los Angeles, Calif.) . Isolation of secretory IgA and free secretory component. The fractionation of secretory IgA and free secretory component from colostrum, parotid saliva, and nasal secretions was done by
127
128
The thickness of the gel layer was approximately 2mm. Small antigen wells were cut in the agarose gel with a tubular cutter and a plastic template. The wells were 3 mm in diameter for secretory IgA and 5 mm for IgE; wells were 20 mm apart. Serial dilutions of a standard solution of secretory IgA or IgE with a known concentration of protein were included in each gel plate. Standard solutions of secretory IgA and IgE. A control solution of secretory IgA was prepared with purified secretory 19A. A control serum with an IgE concentration of approximately 10,000 units was purchased from the same source as the antiserum to IgE (Pharmacia). The exact concentration of this control serum was determined by Dr. K. Ito (University of Tokyo Medical School, Tokyo, Japan) by RSRD (a one-step method) using a standard serum of IgE supplied by the World Health Organization. The IgE concentration of our control serum was 10.15 ug/ml. The IgE reference samples for RSRD were prepared by serial dilution of the control serum with a 1 % solution of bovine serum albumin prepared in borate saline buffer. Results
Total protein concentration and relative concentration of protein fractions of middle ear effusions and sera. Results from these studies are shown in table 1. The mean concentration of total protein in the effusions was higher than that in sera for all patients except those with chronic disease and mucoid effusions. However, these elevations of protein concentration were not statistically significant. A protein fraction of the middle ear effusion corresponding to a fraction in serum in electrophoretic mobility was considered to be the same fraction as that in the serum. Whereas the concentrations of albumin and U1-, ~-, and y-globulin fractions in all the effusions were nearly equal to those in the serum samples, the concentration of the u2-fraction in the effusion fluid was less than that in the sera. Specifically, in all serous effusions from patients with chronic disease, the mean concentration of the u2-fraction was significantly lower than that in sera (P < 0.01). Disc electrophoretic patterns. Twenty-four samples of middle ear effusion and corresponding
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
[8J. The gels were stained with Coomassie brilliant blue. A standard curve was constructed from molecular weights, and assumed weights were assigned to the following proteins and polypeptides: cytochrome c (11,700), chymotrypsinogen (25,700), aldolase (40,000), ovalbumin (43,000), catalase (60,000), serum albumin (68,000), phosphorylase (94,000), and ~-galac tosidase (130,000 daltons). Labeling of the antibodies with 1251. IgG (2 mg) fractionated from antiserum to secretory component by DEAE cellulose chromatography was labeled with 2 mCi of 1251 with use of sodium iodine and chloramin T [9]. Free iodine was removed by dialysis against borate saline buffer solution overnight in a cold chamber (4 C). The 1251-labeled antibodies were mixed with the cold antibodies to secretory component. Five milligrams of 7S v-globulin, fractionated from sheep antiserum to IgE (Pharmacia Fine Chemicals, Uppsala, Sweden) , were labeled with 1251 in the same manner. Immunodiffusion analysis. Double diffusion immunoprecipitation was performed according to the method of Ouchterlony [10J. Immunodiffusion analysis with the 1251-labeled antibodies to secretory component was carried out for detection of secretory component determinants in serous specimens of the middle ear effusion and in sera. To discover the lowest concentration of secretory IgA detectable by radio immunodiffusion analysis, solutions with different concentrations of secretory IgA were placed in the outer wells. The concentrations of secretory IgA solutions were 7.5, 1.88, 0.45, 0.12, 0.03, and 0.0075 mg of secretory IgA/ml. Immunoelectrophoresis. Immunoelectrophoresis [10] with the antiserum to secretory IgA and radioimmunoelectrophoresis with the 1251-labeled antibodies to secretory component were performed on serous and mucoid middle ear effusions and sera for detection of secretory IgA and free secretory component. Radioactive single radial diffusion. Radioactive single radial diffusion (RSRD) as outlined by Mancini [11] was adopted for analysis of secretory IgA and IgE quantitatively in middle ear effusions and sera. The antibody-containing agarose gel was prepared by a mixture of 1.2 % molten agarose gel and the 1251-labeled antibodies.
Mogi et al.
129
Otitis Media with with Effusion Otitis
~
8
~~:~t; ~ ;;;-MO"';OM ~
"=' .2 u .2 ~
~ ~ :; +1 +1 +1 +1 +1 § CIl ''''0\'''('.10\ Z 'S!11ilca: 1'-00",;o\or-:.
~
fI
'-'
~
~
V)
.....
~.-.
~
(I)
§ ~
~
col'
"So
1 Ie (I)
...,
-
...
rIO
0
o
§
~
~
..s
I'" :::
'"" I''-'
u
f'f'\
:6 "1 C!
(I)
.~
"'"o
til
Il.t
~
~
'" '" '" ........ Vi ..... N
Q,
(I) rIO
til
~
'"
(I)
"='
e
;
«
o'"
I:::
(I)
CIl
~
~0l~0lr--:0l1 ~ I'- ~ """ 0\ ;: N .9
~
.;:
,.§
rIO
6 00-1'-""""'01 00 .~ r-: 0'0 ..t r-: 0\ ; ~ >
.s....
~...
o
til
1'-00 'w:t ..... .... -.:t "'""", 00 0 ("-.t
(I)
8 o..to"';",;",;
~
'8 .9
(1") \\0 IVi n IV') nM Ooq-..... 'l:t_o\lnlnM""!0l"1"1"l""!
--. W
.~
"=' ...,
a ·13
I::: ._ o ..r:: ...0
.9
rIO
.2 8
tEl
"'"o
i
U
0.. fIl
C
o
§ C'i -.:t ,...; ,...; ff'i M t;l
~ ~I+1 ~I+1 +1 ti+1 71+1 71+1 i!+1 ~ ~~~~~~ ~~~:;~~ e E In '"
(I)
~
'01'-0\'0"""
e
fl ~:::!:.
~
".0
3~ 11
J:l_
>
'CD
V).......... '"""
~
500 ng/rnl. In only two (2.1 %) of 96 cases did the 19B concentration in effusions exceed both 500 ng/ml and the 19B concentration in the corresponding sample of serum. Two of three serous effusions and all four mucoid effusions that were found to contain > 500 ng of IgE/ml were obtained from patients with the chronic type of disease. There was good correlation (P < 0.01) between values of the effusions and sera in each category, except for those from the group with acute mucoid otitis. Table 5 shows mean levels of IgE in various age groups. No significant differences were noted among these values.
134
structure to secretory IgA found in other external secretions, such as saliva, nasal secretion, and colostrum. In addition, immunodiffusion and immunoelectrophoretic studies in the present investigation showed that secretory IgA isolated from middle ear effusions appears to be antigenically identical to that from other external secretions. Secretory component characterizing secretory IgA is a nonimmunoglobulin glycoprotein that exists in either a bound state (associated with IgA) or a free state (unassociated with IgA) in colostrum and saliva [6, 22]. In this study free secretory component was detected in one (2 % ) of 47 serous effusions and six (20%) of 30 mucoid effusions. This finding indicates that the greater part of the secretory component present in middle ear effusions existed in the bound state. Therefore, the concentration of secretory component determinants in effusions and sera estimated by RSRD with use of 125I-Iabeled antibody to secretory component may represent the concentration of secretory IgA. The results demonstrated that the concentration of secretory IgA in the serous effusions was eight times that in the sera from the corresponding patients, whereas the concentration in mucoid samples was approximately 14 times that in sera; the calculated percentage of secretory IgA in the total IgA of middle ear effusions (both serous and mucoid) was 11 %12 %. This evidence indicates that all middle ear effusions contain appreciable amounts of the secretory product, although the transudate from serum is the major component of effusion. IgE is a more recently defined immunoglobulin containing reaginic antibodies, and recent studies [30-34] have demonstrated elevated levels of IgE in sera and secretions from patients with various atopic allergic diseases. Even though allergic patients have been clinically observed [3539] to be prone to develop otitis media with effusion, the allergic etiology of this disease is still controversial. The controversy results from a lack of direct scientific evidence to support the supposition that the secretory fluid is an allergic effusion. Ishikawa et al. [40] found IgE in two (9.5%) of 21 middle ear effusions obtained from patients four to nine years old. Ogra et al. [28] reported that small amounts of IgE were detected in five (25%) of 20 middle ear effusions from subjects six to 14 years old. With an immuno-
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
that the enhancement of this secretory activity is an important factor contributing to the accumulation of fluid in the middle ear cavity in otitis media with effusion. If middle ear effusion is a resultant product of secretory elements of the membrane lining the middle ear, protein components of the effusion may differ, at least partially, from those of serum. However, results of several investigations [17-20] demonstrated no qualitative difference between the protein components of serum and those of middle ear effusion. Although the number of cases subjected to electrophoretic analysis in this study was small, it was shown that the relative concentration of the u2-fraction in middle ear effusion was significantly lower than that in sera. Similar findings have been reported previously [17, 18, 21]. Results of the disc electrophoretic analysis indicated that this low concentration of the u2-fraction in effusions was due to the absence or very low concentration of fractions of haptoglobin and u2-macroglobuIin. Since immunoglobulins are separated as a diffuse background by disc electrophoresis, evaluation of immunoglobulin fractions by disc electrophoresis is not adequate. Findings from cellulose acetate and disc electrophoretic analyses in this study suggested that the proteins, except for immunoglobulins in the effusions, for the most part come from the serum. Secretory IgA is synthesized locally in the mucous membrane or glandular epithelium [22], and this class of antibody contributes significantly to the immunological, mucosal resistance mechanism, since antibodies to viral and bacterial antigens have been demonstrated in secretory IgA antibody [22-25]. Bernstein et al. [26], Howei et al. [27], and Ogra et al. [28] have demonstrated the existence of secretory IgA in middle ear effusions and the mucosal lining of the middle ear cavity. This evidence suggested that the middle ear cavity takes part in the local immunologic defense mechanism. In the present study, we isolatedsecretory IgA from pooled middle ear effusions and studied and compared its subunit structure with that of other external secretions. Although in this study J-chain, which was recently detected in molecules of secretory IgA by Halpern and Koshland [29] and Mestecky et al. [5], was not identified, the secretory IgA isolated from middle ear effusions was very similar in subunit
Mogi et al.
Otitis Media with Effusion
References
1. Senturia, B. H. Classification of middle ear effusions. Ann. Otol. Rhinol. Laryngol, 79:358-370, 1970. 2. Kohn, J. Cellulose acetate electrophoresis and immunodiffusion techniques. In I. Smith [ed.]. Chromatographic and electrophoretic techniques. Vol. 2 (Zone electrophoresis). 1st ed. William Heinemann Medical Books, London, 1960, p. 55-90.
3. Davis, R J. Disc electrophoresis. II. Method and application to human serum proteins. Ann. N.Y. Acad. Sci. 121:404-427, 1964. 4. Tomasi, T. B., Jr., Tan, E. M., Solomon, A., Prendergast, R. A. Characteristics of an immune system common to certain external secretions. J. Exp. Med. 121:101-124, 1965. 5. Mestecky, J., Kulhavy, R., Kraus, F. W. Studies on human secretory immunoglobulin A. II. Subunit structure. J. Immunol. 108:738-747, 1972. 6. Kobayashi, K Secretory IgA. Physico-Chemical Biology (Japan) 18: 107-110, 1974. 7. Wofsy, L., Burr, R The use of affinity chromatography for the specific purification of antibodies and antigens. J. Immunol. 103: 380-382, 1969. 8. Weber, K, Osborn, M. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. BioI. Chern. 244:4406-4412, 1969. 9. Talmage, D. W., Claman, H. N. External labeling of proteins with Jl3l, Jl25 and Jl30. In C. A. Williams and M. W. Chase [ed.]. Methods in immunology and immunochemistry. Vol. 1. Academic Press, New York, 1967, p. 389-391. 10. Ouchterlony, O. Handbook of immunodiffusion and immunoelectrophoresis. Ann Arbor Science Publishers, Ann Arbor, 1968. 215 p. 11. Mancini, G., Carbonara, A. 0., Heremans, J. F. Immunochemical quancitation of antigens by single radial immunodiffusion. Immunochemistry 2: 235-254, 1965. 12. Bienenstock, J., Tourville, D., Tomasi, T. R, Jr. The secretion of immunoglobulins by the human salivary glands. In S. Y. Botelho, F. P. Brooks, and W. R Shelly [ed.]. The exocrine glands. University of Pennsylvania Press, Philadelphia, 1969, p. 187-198. 13. Sade, J. Pathology and pathogenesis of serous otitis media. Arch. Otolaryngol, 84:297-305, 1966. 14. Hentzer, E. Ultrastructure of the normal mucosa in the human middle ear, mastoid cavities, and Eustachian tube. Ann. Oto1. Rhino1. Laryngo1. 79:1143-1157, 1970. 15. Lim, D. J., Shimada, T. Secretory activity of normal middle ear epithelium. Scanning and transmission electron microscopic observation. Ann. Oto1. Rhinol. Laryngol. 80:319-329, 1971. 16. Lim, D. J., Birck, H. Ultrastructural pathology of the middle ear mucosa in serous otitis media. Ann. Oto1. Rhinol. Laryngol, 80:838-853, 1973. 17. Carlson, L. A., Lokk, T. Protein studies of transudates of the middle ear. Scand. J. Clin. Lab. Invest. 7:43-48, 1955. 18. Juhn, S. K., Huff, J. S., Paparella, M. M. Biochemical analysis of middle ear effusions. Preliminary report. Ann. Otol. Rhinol. Laryngol. 80:347-353, 1971. 19. Tonder, 0., Gundersen, T. Nature of the fluid in serous otitis media. Arch. Otolaryngo1. 93 :473478, 1971.
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
fluorescence study, Ogra et al. failed to find 19B in the middle ear mucosa of patients diagnosed as having otitis media with effusion [28]. Results of the estimation of 19B level suggested that 19B in middle ear effusions, rather than being a local product, may be derived from the serum. However, there were a few cases in which the concentration of 19B in the effusion exceeded both 500 ng/rnl and the 19B concentration in the corresponding sample of serum. These effusions might be an atopic allergic fluid. We [41] investigated 19B antibody to mites in effusions and sera with a radioimmunoallergosorbent test and found antibody in five (8.9%) of 56 middle ear effusions. It should be considered that most of the patients in the present study were elderly. In older patients allergic attacks are less frequent and less severe [36]. Moreover, allergy has been implicated in otitis media with effusion more frequently in children than in adult patients [35-39]. Thus if the 19B concentration is estimated from selected specimens from young patients with recurrent episodes of otitis media with effusion, the data may be different from those in the present study. In fact, Phillips et al. [42] recently found the mean concentration of 19B in 26 middle ear effusions collected from children (aged two to 12 years) to be 3,026 ng/rnl. The findings of this immunochemical study on otitis media with effusion suggest that, while serous middle ear effusions are a mixture of normal ear secretions that bathe the membrane surface and transudates from serum, mucoid effusions are the result of enhanced epithelial secretory activity and transudates. It was confirmed that there is a local immunity in the middle ear cavity of patients who have otitis media with effusion. The results of estimation of 19B concentration suggested to us that 19B in middle ear effusions obtained from old patients, rather than being a local product, may be derived from the serum.
135
136
30. Johansson, S. G. O. 19B in allergic diseases. Proc. R. Soc. Med. 62:975-976, 1969. 31. Smith, H. J., Ozkaragoz, K., Gokcen, M. A simplified radioimmunoassay technique for measuring human IgE. J. Allergy Clin. Immunol. 50: 193-197, 1972. 32. Yunginger, J. W., Gleich, G. J. Seasonal changes in serum and nasal IgE concentrations. J. Allergy Clin. Immunol. 47:87, 1971. 33. Tse, K. S., Wicher, K., Arbesman, C. E. IgE antibodies in nasal secretions of ragweed-allergic subjects. J. Allergy Clin. Immunol. 46:352-358, 1970. 34. Ishizaka, K., Newcomb, R. W. Presence of yE in nasal washings and sputum from asthmatic patients. J. Allergy Clin. Immunol. 46: 197-204, 1970. 35. Jordan, R. Chronic secretory otitis media. Laryngoscope 59: 1002-1015, 1949. 36. Solow, I. A. Is serous otitis media due to allergy or infection? Ann. Allergy 16:297-299, 1958. 37. Leeks, H. I. Allergic aspects of serous otitis media in childhood. N.Y. State J. Med. 61:2737-2743, 1961. 38. McGovern, J. P., Haywood, T. J., Fernandez, A. A. Allergy and secretory otitis media. An analysis of 512 cases. J.A.M.A. 200: 124-128, 1967. 39. Viscomi, G. J. Allergic secretory otitis media: an approach to management. Laryngoscope 85:751758, 1975. 40. Ishikawa, T., Bernstein, J., Reisman, R. E., Arhesman, C. E. Secretory otitis media: immunologic studies of middle ear secretions. J. Allergy Clin. Immunol. 50:319-325, 1972. 41. Mogi, G., Maeda, S., Yoshida, T., Watanabe, N. Radioimmunoassay of IgE studies on middle ear effusions. Acta Otolaryngol., 1976 (in press). 42. Phillips, M. J., Knight, N. J., Manning, H., Abbott, A. L. IgE and secretory otitis media. Lancet 2: 1176-1178, 1975.
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
20. Aro, M. J. T., Kouvalainen, K. Proteins of middle ear secretion in serous otitis. An electrophoretic and immunoelectrophoretic study. Acta Otolaryngol. [Suppl.] (Stockh.) 224:385-389, 1967. 21. Bernstein, J. M., Tomasi, T. B., Jr., Ogra, P. The immunochemistry of middle ear effusions. Arch. Otolaryngol. 99:320-326, 1974. 22. Tomasi, T. B., Jr., Bienenstock, J. Secretory immunoglobulins. Adv, Immunol. 9: 1-96, 1968. 23. Adinolfi, M., Glynn, A. A., Lindsay, M., Milne, C. M. Serological properties of y A antibodies to Escherichia coli present in human colostrum. Immunology 10:517-526, 1966. 24. Douglas, R. G., Jr., Rossen, R. D., Butler, W. T., Cough, R. B. Rhinovirus neutralizing antibody in tears, parotid saliva, nasal secretions and serum. J. Immunol. 99:297-303, 1967. 25. Waldman, R. H., Jurgensen, P. F., Olsen, G. N., Ganguly, R., Johnson, J. E., III. Immune response of the human respiratory tract. I. Immunoglobulin levels and influenza virus vaccine antibody response. J. Immunol. 111:38-41, 1973. 26. Bernstein, J. M., Hayes, E. R., Ishikawa, T., Tomasi, T. B., Jr., Herd, J. K. Secretory otitis media: a histopathologic and immunochemical report. Trans. Am. Acad. Ophthalmol. Otolaryngol. 76: 1305-1318, 1972. 27. Howie, V. M., Ploussard, J. H., Sloyer, J. L., Johnson, R. B., Jr. Immunoglobulins of the middle ear fluid in acute otitis media: relationship to serum immunoglobulin concentrations and bacterial cultures. Infec. Immun. 7:589-593, 1973. 28. Ogra, P. L., Bernstein, J. M., Yurchak, A. M., Coppola, P. R., Tomasi, T. B., Jr. Characteristics of secretory immune system in human middle ear: implications in otitis media. J. Immunol. 112: 488-495, 1974. 29. Halpern, M. S., Koshland, M. E. Novel subunit in secretory IgA. Nature 228: 1276-1278, 1970.
Mogi et al.
Immunochemistry of Otitis Media with Effusion From the Department of Otolaryngology, Yamaguchi University School of Medicine, Ube, Yamaguchi-ken, Japan
Goro Mogi, Shoichi Maeda, Toyoharu Yoshida, and Noritake Watanabe
Within recent years, striking biochemical and immunochemical analytic techniques have been developed and have opened a new approach to the study of otitis media with effusion. This study was designed to examine immunoglobulin components of middle ear effusions, particularly secretory immunoglobulin A, for clarification of the nature of fluid in the middle ear cavity and for investigation of the existence of a local immune system in the middle ear.
when the middle ear effusion was aspirated. These patients were between six and 78 years of age; 50% were> 60 years old, 35% were between 17 and 59 years old, and 15 % were < 16 years old. Middle ear effusions were classified as serous type or mucoid type by gross appearance; clear fluid with a low viscosity similar to that of serum was referred to as serous, whereas viscous fluid was referred to as mucoid type. Our mucoid type involved the mucopurulent type classified by Senturia [1]. This type of effusion was found in seven cases, which were subjected to only qualitative analysis of secretory IgA. The aspirates were also broken down by origin from an acute case or from a chronic case. Acute cases had short duration of illness (less than three weeks) and transient existence of the effusion, whereas chronic cases had long duration of illness (more than three weeks) and recurrent existence of the effusion. The amount of ear aspirate varied among patients, but it was usually between 0.1 and 0.6 ml, All specimens of middle ear effusion and serum were kept in a deep freezer (-20 C) until use. Once the specimen had been melted to use for one of the procedures in the study, this sample was not used in any other procedure except affinity chromatography. Therefore, each procedure was usually performed upon one aspirate.
Materials and Methods
Middle ear effusions were collected by aspiration through the eardrums of more than 400 patients diagnosed as having otitis media with effusion. In all cases, acute signs of otitis media, such as earache and hyperemia of the eardrum, were absent
Received for publication March 12, 1975, and in revised form October 1, 1975. This work was supported by grant no. 848193 from the Ministry of Education, Japan. We wish to thank Drs. T. Nakashima, M. Ogata, Y. Tawara, and Y. Onaka for providing specimens. Please address requests for reprints to Dr. Goro Mogi, Department of Otolaryngology, Yamaguchi University School of Medicine, Ube, Yamaguchi-ken, Japan.
126
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
For investigation of the nature and origin of middle ear effusions, immunochemical studies were performed on more than 400 patients diagnosed as having otitis media with effusion. Although results of cellulose acetate and disc electrophoretic analyses and quantitation of IgG, IgA, and IgM suggested that proteins found in the effusions were derived for the most part from the serum, quantitative analysis of secretory IgA revealed the existence of appreciable amounts of secretory IgA in both serous and mucoid effusions. The antigenicity and subunit structure of the secretory IgA isolated from middle ear effusions were identical or very similar to those of secretory IgA obtained from other external secretions. Radioactive single radial diffusion analysis of IgE showed that the mean concentrations of IgE in effusions and sera were within normal ranges. Findings of this study suggest that the middle ear maintains the local immunologic defense system, that the middle ear effusion is at least partially an external secretion, and that IgE in middle ear effusions obtained from old patients, rather than being a local product, may be derived from the serum.
Otitis Media with Effusion
methods described by Tomasi et al. [4], Mestecky et al. [5], and Kobayashi [6]. Antisera to secretory IgA, free secretory component, and a-chain. Production of antiserum to human secretory IgA was induced by immunization of rabbits with isolated secretory IgA. The rabbit antiserum to secretory IgA was absorbed with cord serum to make mono specific antiserum. Antiserum to secretory component was made by absorption of the antiserum to secretory IgA with pooled normal human sera and by immunization of rabbits with free secretory component. Rabbit antiserum to a-chain was obtained from Behringwerke, Marburg-Lahn, Germany. Preparation of 7S IgA. The 7S IgA (serum IgA) was purified from the serum of a patient with IgA myeloma. The crude y-globulins were precipitated with (NH4) 2S04 at 40% final saturation. The precipitate was dissolved again and dialyzed against 0.01 M phosphate buffer at pH 7.5. The crude y-globulins were subjected to ionexchange chromatography on DEAE cellulose. Eluates in 0.1 M phosphate buffer (pH 6.2) were concentrated, dialyzed against saline buffered with Tris-HCI (pH 7.4), and filtered through Seph arose 6B gel packed in a 1.5- X 100-cm column. The main fraction, appearing as the second peak, was rich in 7S IgA. Isolation of secretory IgA from middle ear effusions. Affinity chromatography was used to isolate secretory IgA from the pooled middle ear effusion. This technique was performed as described by Wofsy and Burr [7]. Sepharose 6B gel coupled to antibodies to secretory component was packed to a height of 8 em in a column (1.5X 30-cm), and 10 ml of the pooled middle ear effusion was applied. After nonspecific proteins were washed out, eluates were adjusted to neutral pH with 0.17 M glycine-Hel buffer (pH 2.3), concentrated, and passed through Sepharose 6B gel. The fraction containing secretory IgA was collected and concentrated. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis. For investigation of the constituents of the secretory IgA obtained from middle ear effusions and other external secretions, and for determination of the molecular weights of these polypeptide chains, sodium dodecyl sulfate (SDS)polyacrylamide gel electrophoresis was performed according to the method of Weber and Osborn
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
A blood sample for serum was taken from each patient at the same time that the effusion was aspirated. Human colostrum was provided by the Department of Gynecology and Obstetrics, Yamaguchi University School of Medicine. Parotid secretions were collected from normal adult humans by insertion of a capillary tube into the Stensen's ducts. Nasal secretions were collected from patients with chronic rhinitis or chronic paranasal sinus infection. These secretions were the source of secretory IgA and free secretory component. Determination of total protein concentration. The total concentration of protein was measured by a Hitachi serum protein meter (refractometer) (Tsukasa Kogyu, Nikko, Ibaragi, Japan). Quantitative analysis of protein fractions. Microzone electrophoresis on cellulose acetate (Separaxw; Joko Sangyo, Co., Ltd., Bunkyo-ku, Tokyo, Japan) and densitometry were used for the quantitative analysis. Cellulose acetate electrophoresis was performed as outlined by Kohn [2] with use of 0.07 M Veronal buffer at pH 8.6. The quantitative determination of the separate components, which were stained with Ponceau's 3R (Matheson Co., Elk Grove Village, Ill.), was performed with use of an automatic recording and integrating scanner (Kayagaki Instruments, Bunkyo-ku, Tokyo, Japan). Disc electrophoresis. Disc electrophoresis was performed according to the method of Davis [3] in 7.5% polyacrylamide gel equilibrated with Tris-HCI buffer (pH 8.9) with a constant current of 2 rnA per tube for 90 min. Samples of 3 or 4 III of middle ear effusion or sera were used in each tube. After the completion of electrophoresis, the gels were removed from the glass tubes, placed into a staining solution of 1% amido Schwarz® dye (E. Merck, Darmstadt, B.R.-Germany) in 7% acetic acid, and washed with 7 % acetic acid to remove excess dye. Quantitative analysis of IgG, IgA, and IgM. Levels of IgG, IgA, and IgM in both middle ear effusion and sera were determined with Hyland immunoplates (Hyland Laboratories, Los Angeles, Calif.) . Isolation of secretory IgA and free secretory component. The fractionation of secretory IgA and free secretory component from colostrum, parotid saliva, and nasal secretions was done by
127
128
The thickness of the gel layer was approximately 2mm. Small antigen wells were cut in the agarose gel with a tubular cutter and a plastic template. The wells were 3 mm in diameter for secretory IgA and 5 mm for IgE; wells were 20 mm apart. Serial dilutions of a standard solution of secretory IgA or IgE with a known concentration of protein were included in each gel plate. Standard solutions of secretory IgA and IgE. A control solution of secretory IgA was prepared with purified secretory 19A. A control serum with an IgE concentration of approximately 10,000 units was purchased from the same source as the antiserum to IgE (Pharmacia). The exact concentration of this control serum was determined by Dr. K. Ito (University of Tokyo Medical School, Tokyo, Japan) by RSRD (a one-step method) using a standard serum of IgE supplied by the World Health Organization. The IgE concentration of our control serum was 10.15 ug/ml. The IgE reference samples for RSRD were prepared by serial dilution of the control serum with a 1 % solution of bovine serum albumin prepared in borate saline buffer. Results
Total protein concentration and relative concentration of protein fractions of middle ear effusions and sera. Results from these studies are shown in table 1. The mean concentration of total protein in the effusions was higher than that in sera for all patients except those with chronic disease and mucoid effusions. However, these elevations of protein concentration were not statistically significant. A protein fraction of the middle ear effusion corresponding to a fraction in serum in electrophoretic mobility was considered to be the same fraction as that in the serum. Whereas the concentrations of albumin and U1-, ~-, and y-globulin fractions in all the effusions were nearly equal to those in the serum samples, the concentration of the u2-fraction in the effusion fluid was less than that in the sera. Specifically, in all serous effusions from patients with chronic disease, the mean concentration of the u2-fraction was significantly lower than that in sera (P < 0.01). Disc electrophoretic patterns. Twenty-four samples of middle ear effusion and corresponding
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
[8J. The gels were stained with Coomassie brilliant blue. A standard curve was constructed from molecular weights, and assumed weights were assigned to the following proteins and polypeptides: cytochrome c (11,700), chymotrypsinogen (25,700), aldolase (40,000), ovalbumin (43,000), catalase (60,000), serum albumin (68,000), phosphorylase (94,000), and ~-galac tosidase (130,000 daltons). Labeling of the antibodies with 1251. IgG (2 mg) fractionated from antiserum to secretory component by DEAE cellulose chromatography was labeled with 2 mCi of 1251 with use of sodium iodine and chloramin T [9]. Free iodine was removed by dialysis against borate saline buffer solution overnight in a cold chamber (4 C). The 1251-labeled antibodies were mixed with the cold antibodies to secretory component. Five milligrams of 7S v-globulin, fractionated from sheep antiserum to IgE (Pharmacia Fine Chemicals, Uppsala, Sweden) , were labeled with 1251 in the same manner. Immunodiffusion analysis. Double diffusion immunoprecipitation was performed according to the method of Ouchterlony [10J. Immunodiffusion analysis with the 1251-labeled antibodies to secretory component was carried out for detection of secretory component determinants in serous specimens of the middle ear effusion and in sera. To discover the lowest concentration of secretory IgA detectable by radio immunodiffusion analysis, solutions with different concentrations of secretory IgA were placed in the outer wells. The concentrations of secretory IgA solutions were 7.5, 1.88, 0.45, 0.12, 0.03, and 0.0075 mg of secretory IgA/ml. Immunoelectrophoresis. Immunoelectrophoresis [10] with the antiserum to secretory IgA and radioimmunoelectrophoresis with the 1251-labeled antibodies to secretory component were performed on serous and mucoid middle ear effusions and sera for detection of secretory IgA and free secretory component. Radioactive single radial diffusion. Radioactive single radial diffusion (RSRD) as outlined by Mancini [11] was adopted for analysis of secretory IgA and IgE quantitatively in middle ear effusions and sera. The antibody-containing agarose gel was prepared by a mixture of 1.2 % molten agarose gel and the 1251-labeled antibodies.
Mogi et al.
129
Otitis Media with with Effusion Otitis
~
8
~~:~t; ~ ;;;-MO"';OM ~
"=' .2 u .2 ~
~ ~ :; +1 +1 +1 +1 +1 § CIl ''''0\'''('.10\ Z 'S!11ilca: 1'-00",;o\or-:.
~
fI
'-'
~
~
V)
.....
~.-.
~
(I)
§ ~
~
col'
"So
1 Ie (I)
...,
-
...
rIO
0
o
§
~
~
..s
I'" :::
'"" I''-'
u
f'f'\
:6 "1 C!
(I)
.~
"'"o
til
Il.t
~
~
'" '" '" ........ Vi ..... N
Q,
(I) rIO
til
~
'"
(I)
"='
e
;
«
o'"
I:::
(I)
CIl
~
~0l~0lr--:0l1 ~ I'- ~ """ 0\ ;: N .9
~
.;:
,.§
rIO
6 00-1'-""""'01 00 .~ r-: 0'0 ..t r-: 0\ ; ~ >
.s....
~...
o
til
1'-00 'w:t ..... .... -.:t "'""", 00 0 ("-.t
(I)
8 o..to"';",;",;
~
'8 .9
(1") \\0 IVi n IV') nM Ooq-..... 'l:t_o\lnlnM""!0l"1"1"l""!
--. W
.~
"=' ...,
a ·13
I::: ._ o ..r:: ...0
.9
rIO
.2 8
tEl
"'"o
i
U
0.. fIl
C
o
§ C'i -.:t ,...; ,...; ff'i M t;l
~ ~I+1 ~I+1 +1 ti+1 71+1 71+1 i!+1 ~ ~~~~~~ ~~~:;~~ e E In '"
(I)
~
'01'-0\'0"""
e
fl ~:::!:.
~
".0
3~ 11
J:l_
>
'CD
V).......... '"""
~
500 ng/rnl. In only two (2.1 %) of 96 cases did the 19B concentration in effusions exceed both 500 ng/ml and the 19B concentration in the corresponding sample of serum. Two of three serous effusions and all four mucoid effusions that were found to contain > 500 ng of IgE/ml were obtained from patients with the chronic type of disease. There was good correlation (P < 0.01) between values of the effusions and sera in each category, except for those from the group with acute mucoid otitis. Table 5 shows mean levels of IgE in various age groups. No significant differences were noted among these values.
134
structure to secretory IgA found in other external secretions, such as saliva, nasal secretion, and colostrum. In addition, immunodiffusion and immunoelectrophoretic studies in the present investigation showed that secretory IgA isolated from middle ear effusions appears to be antigenically identical to that from other external secretions. Secretory component characterizing secretory IgA is a nonimmunoglobulin glycoprotein that exists in either a bound state (associated with IgA) or a free state (unassociated with IgA) in colostrum and saliva [6, 22]. In this study free secretory component was detected in one (2 % ) of 47 serous effusions and six (20%) of 30 mucoid effusions. This finding indicates that the greater part of the secretory component present in middle ear effusions existed in the bound state. Therefore, the concentration of secretory component determinants in effusions and sera estimated by RSRD with use of 125I-Iabeled antibody to secretory component may represent the concentration of secretory IgA. The results demonstrated that the concentration of secretory IgA in the serous effusions was eight times that in the sera from the corresponding patients, whereas the concentration in mucoid samples was approximately 14 times that in sera; the calculated percentage of secretory IgA in the total IgA of middle ear effusions (both serous and mucoid) was 11 %12 %. This evidence indicates that all middle ear effusions contain appreciable amounts of the secretory product, although the transudate from serum is the major component of effusion. IgE is a more recently defined immunoglobulin containing reaginic antibodies, and recent studies [30-34] have demonstrated elevated levels of IgE in sera and secretions from patients with various atopic allergic diseases. Even though allergic patients have been clinically observed [3539] to be prone to develop otitis media with effusion, the allergic etiology of this disease is still controversial. The controversy results from a lack of direct scientific evidence to support the supposition that the secretory fluid is an allergic effusion. Ishikawa et al. [40] found IgE in two (9.5%) of 21 middle ear effusions obtained from patients four to nine years old. Ogra et al. [28] reported that small amounts of IgE were detected in five (25%) of 20 middle ear effusions from subjects six to 14 years old. With an immuno-
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
that the enhancement of this secretory activity is an important factor contributing to the accumulation of fluid in the middle ear cavity in otitis media with effusion. If middle ear effusion is a resultant product of secretory elements of the membrane lining the middle ear, protein components of the effusion may differ, at least partially, from those of serum. However, results of several investigations [17-20] demonstrated no qualitative difference between the protein components of serum and those of middle ear effusion. Although the number of cases subjected to electrophoretic analysis in this study was small, it was shown that the relative concentration of the u2-fraction in middle ear effusion was significantly lower than that in sera. Similar findings have been reported previously [17, 18, 21]. Results of the disc electrophoretic analysis indicated that this low concentration of the u2-fraction in effusions was due to the absence or very low concentration of fractions of haptoglobin and u2-macroglobuIin. Since immunoglobulins are separated as a diffuse background by disc electrophoresis, evaluation of immunoglobulin fractions by disc electrophoresis is not adequate. Findings from cellulose acetate and disc electrophoretic analyses in this study suggested that the proteins, except for immunoglobulins in the effusions, for the most part come from the serum. Secretory IgA is synthesized locally in the mucous membrane or glandular epithelium [22], and this class of antibody contributes significantly to the immunological, mucosal resistance mechanism, since antibodies to viral and bacterial antigens have been demonstrated in secretory IgA antibody [22-25]. Bernstein et al. [26], Howei et al. [27], and Ogra et al. [28] have demonstrated the existence of secretory IgA in middle ear effusions and the mucosal lining of the middle ear cavity. This evidence suggested that the middle ear cavity takes part in the local immunologic defense mechanism. In the present study, we isolatedsecretory IgA from pooled middle ear effusions and studied and compared its subunit structure with that of other external secretions. Although in this study J-chain, which was recently detected in molecules of secretory IgA by Halpern and Koshland [29] and Mestecky et al. [5], was not identified, the secretory IgA isolated from middle ear effusions was very similar in subunit
Mogi et al.
Otitis Media with Effusion
References
1. Senturia, B. H. Classification of middle ear effusions. Ann. Otol. Rhinol. Laryngol, 79:358-370, 1970. 2. Kohn, J. Cellulose acetate electrophoresis and immunodiffusion techniques. In I. Smith [ed.]. Chromatographic and electrophoretic techniques. Vol. 2 (Zone electrophoresis). 1st ed. William Heinemann Medical Books, London, 1960, p. 55-90.
3. Davis, R J. Disc electrophoresis. II. Method and application to human serum proteins. Ann. N.Y. Acad. Sci. 121:404-427, 1964. 4. Tomasi, T. B., Jr., Tan, E. M., Solomon, A., Prendergast, R. A. Characteristics of an immune system common to certain external secretions. J. Exp. Med. 121:101-124, 1965. 5. Mestecky, J., Kulhavy, R., Kraus, F. W. Studies on human secretory immunoglobulin A. II. Subunit structure. J. Immunol. 108:738-747, 1972. 6. Kobayashi, K Secretory IgA. Physico-Chemical Biology (Japan) 18: 107-110, 1974. 7. Wofsy, L., Burr, R The use of affinity chromatography for the specific purification of antibodies and antigens. J. Immunol. 103: 380-382, 1969. 8. Weber, K, Osborn, M. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. BioI. Chern. 244:4406-4412, 1969. 9. Talmage, D. W., Claman, H. N. External labeling of proteins with Jl3l, Jl25 and Jl30. In C. A. Williams and M. W. Chase [ed.]. Methods in immunology and immunochemistry. Vol. 1. Academic Press, New York, 1967, p. 389-391. 10. Ouchterlony, O. Handbook of immunodiffusion and immunoelectrophoresis. Ann Arbor Science Publishers, Ann Arbor, 1968. 215 p. 11. Mancini, G., Carbonara, A. 0., Heremans, J. F. Immunochemical quancitation of antigens by single radial immunodiffusion. Immunochemistry 2: 235-254, 1965. 12. Bienenstock, J., Tourville, D., Tomasi, T. R, Jr. The secretion of immunoglobulins by the human salivary glands. In S. Y. Botelho, F. P. Brooks, and W. R Shelly [ed.]. The exocrine glands. University of Pennsylvania Press, Philadelphia, 1969, p. 187-198. 13. Sade, J. Pathology and pathogenesis of serous otitis media. Arch. Otolaryngol, 84:297-305, 1966. 14. Hentzer, E. Ultrastructure of the normal mucosa in the human middle ear, mastoid cavities, and Eustachian tube. Ann. Oto1. Rhino1. Laryngo1. 79:1143-1157, 1970. 15. Lim, D. J., Shimada, T. Secretory activity of normal middle ear epithelium. Scanning and transmission electron microscopic observation. Ann. Oto1. Rhinol. Laryngol. 80:319-329, 1971. 16. Lim, D. J., Birck, H. Ultrastructural pathology of the middle ear mucosa in serous otitis media. Ann. Oto1. Rhinol. Laryngol, 80:838-853, 1973. 17. Carlson, L. A., Lokk, T. Protein studies of transudates of the middle ear. Scand. J. Clin. Lab. Invest. 7:43-48, 1955. 18. Juhn, S. K., Huff, J. S., Paparella, M. M. Biochemical analysis of middle ear effusions. Preliminary report. Ann. Otol. Rhinol. Laryngol. 80:347-353, 1971. 19. Tonder, 0., Gundersen, T. Nature of the fluid in serous otitis media. Arch. Otolaryngo1. 93 :473478, 1971.
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
fluorescence study, Ogra et al. failed to find 19B in the middle ear mucosa of patients diagnosed as having otitis media with effusion [28]. Results of the estimation of 19B level suggested that 19B in middle ear effusions, rather than being a local product, may be derived from the serum. However, there were a few cases in which the concentration of 19B in the effusion exceeded both 500 ng/rnl and the 19B concentration in the corresponding sample of serum. These effusions might be an atopic allergic fluid. We [41] investigated 19B antibody to mites in effusions and sera with a radioimmunoallergosorbent test and found antibody in five (8.9%) of 56 middle ear effusions. It should be considered that most of the patients in the present study were elderly. In older patients allergic attacks are less frequent and less severe [36]. Moreover, allergy has been implicated in otitis media with effusion more frequently in children than in adult patients [35-39]. Thus if the 19B concentration is estimated from selected specimens from young patients with recurrent episodes of otitis media with effusion, the data may be different from those in the present study. In fact, Phillips et al. [42] recently found the mean concentration of 19B in 26 middle ear effusions collected from children (aged two to 12 years) to be 3,026 ng/rnl. The findings of this immunochemical study on otitis media with effusion suggest that, while serous middle ear effusions are a mixture of normal ear secretions that bathe the membrane surface and transudates from serum, mucoid effusions are the result of enhanced epithelial secretory activity and transudates. It was confirmed that there is a local immunity in the middle ear cavity of patients who have otitis media with effusion. The results of estimation of 19B concentration suggested to us that 19B in middle ear effusions obtained from old patients, rather than being a local product, may be derived from the serum.
135
136
30. Johansson, S. G. O. 19B in allergic diseases. Proc. R. Soc. Med. 62:975-976, 1969. 31. Smith, H. J., Ozkaragoz, K., Gokcen, M. A simplified radioimmunoassay technique for measuring human IgE. J. Allergy Clin. Immunol. 50: 193-197, 1972. 32. Yunginger, J. W., Gleich, G. J. Seasonal changes in serum and nasal IgE concentrations. J. Allergy Clin. Immunol. 47:87, 1971. 33. Tse, K. S., Wicher, K., Arbesman, C. E. IgE antibodies in nasal secretions of ragweed-allergic subjects. J. Allergy Clin. Immunol. 46:352-358, 1970. 34. Ishizaka, K., Newcomb, R. W. Presence of yE in nasal washings and sputum from asthmatic patients. J. Allergy Clin. Immunol. 46: 197-204, 1970. 35. Jordan, R. Chronic secretory otitis media. Laryngoscope 59: 1002-1015, 1949. 36. Solow, I. A. Is serous otitis media due to allergy or infection? Ann. Allergy 16:297-299, 1958. 37. Leeks, H. I. Allergic aspects of serous otitis media in childhood. N.Y. State J. Med. 61:2737-2743, 1961. 38. McGovern, J. P., Haywood, T. J., Fernandez, A. A. Allergy and secretory otitis media. An analysis of 512 cases. J.A.M.A. 200: 124-128, 1967. 39. Viscomi, G. J. Allergic secretory otitis media: an approach to management. Laryngoscope 85:751758, 1975. 40. Ishikawa, T., Bernstein, J., Reisman, R. E., Arhesman, C. E. Secretory otitis media: immunologic studies of middle ear secretions. J. Allergy Clin. Immunol. 50:319-325, 1972. 41. Mogi, G., Maeda, S., Yoshida, T., Watanabe, N. Radioimmunoassay of IgE studies on middle ear effusions. Acta Otolaryngol., 1976 (in press). 42. Phillips, M. J., Knight, N. J., Manning, H., Abbott, A. L. IgE and secretory otitis media. Lancet 2: 1176-1178, 1975.
Downloaded from http://jid.oxfordjournals.org/ at University of Lethbridge on September 11, 2015
20. Aro, M. J. T., Kouvalainen, K. Proteins of middle ear secretion in serous otitis. An electrophoretic and immunoelectrophoretic study. Acta Otolaryngol. [Suppl.] (Stockh.) 224:385-389, 1967. 21. Bernstein, J. M., Tomasi, T. B., Jr., Ogra, P. The immunochemistry of middle ear effusions. Arch. Otolaryngol. 99:320-326, 1974. 22. Tomasi, T. B., Jr., Bienenstock, J. Secretory immunoglobulins. Adv, Immunol. 9: 1-96, 1968. 23. Adinolfi, M., Glynn, A. A., Lindsay, M., Milne, C. M. Serological properties of y A antibodies to Escherichia coli present in human colostrum. Immunology 10:517-526, 1966. 24. Douglas, R. G., Jr., Rossen, R. D., Butler, W. T., Cough, R. B. Rhinovirus neutralizing antibody in tears, parotid saliva, nasal secretions and serum. J. Immunol. 99:297-303, 1967. 25. Waldman, R. H., Jurgensen, P. F., Olsen, G. N., Ganguly, R., Johnson, J. E., III. Immune response of the human respiratory tract. I. Immunoglobulin levels and influenza virus vaccine antibody response. J. Immunol. 111:38-41, 1973. 26. Bernstein, J. M., Hayes, E. R., Ishikawa, T., Tomasi, T. B., Jr., Herd, J. K. Secretory otitis media: a histopathologic and immunochemical report. Trans. Am. Acad. Ophthalmol. Otolaryngol. 76: 1305-1318, 1972. 27. Howie, V. M., Ploussard, J. H., Sloyer, J. L., Johnson, R. B., Jr. Immunoglobulins of the middle ear fluid in acute otitis media: relationship to serum immunoglobulin concentrations and bacterial cultures. Infec. Immun. 7:589-593, 1973. 28. Ogra, P. L., Bernstein, J. M., Yurchak, A. M., Coppola, P. R., Tomasi, T. B., Jr. Characteristics of secretory immune system in human middle ear: implications in otitis media. J. Immunol. 112: 488-495, 1974. 29. Halpern, M. S., Koshland, M. E. Novel subunit in secretory IgA. Nature 228: 1276-1278, 1970.
Mogi et al.
