On the mechanisms of transport and biological significance of antibodies in external secretions.код для вставкиСкачать
CURRENT COMMENT On the Mechanisms of Transport and Biological Significance of Antibodies in External Secretions By THOMAS B. TOMASI, JR., M.D., b . D . Recent evidence suggests that the antibodies present in certain external secretions are produced locally, and that the predominant immunoglobulin in these secretions appears to be yA, which is complexed to a nonimmunoglobulin glycoprotein referred to as the secretory “piece” or T chain. The secretory im- A munoglobulin system may be of importance in determining local flora and the resistance of mucous membranes to colonization by potentially pathogenic bacteria and viruses, and may be involved in hypersensitivity reactions on mucous surfaces as well as in certain so-called autoimmune processes. of observations in the older literature indicate that under certain circumstances susceptibility to infection and resistance following immunization is not directly related to the titer of serum antibody. Studies, particularly those of Burrows and Havens1 on experimental cholera, suggest that the titers of antibodies present in external secretions such as those of the GI tract are better correlated with resistance to infection than those of serum, and moreover that secretory antibodies are not derived from serum, at least by simple transudation. Recently the characterization and description of the secretory immunoglobulin system: the most outstanding feature of which is the predominance of the yA type of immunoglobulins, have helped to clarify some of these older findings. It has been found that nonvascular fluids can, for the most part, be classified into two groups. Internal secretions, such as the aqueous fluid of the eye, cerebral spinal fluid, and synovial fluid have a yG:yA ratio which is similar to that of serum, i.e., 5:1. External secretions, which bathe mucous membranes which are in direct continuity with the external environment, show a predominance of yA. External secretions include tears, nasal fluid, saliva, respiratory and gastrointestinal tract secretions, and colostrum. It has been found that the yA molecule of secretions has unique chemical as well as antigenic properties and that these characteristics are conferred on the molecule by the presence of a nonimmunoglobulin glycoprotein termed the secretory “piece” or T component.2 A similar type of system probably exists in many species and has been most thoroughly investigated in the rabbit? Evidence has been presented4 that the 11s secretory molecule, having a molecular weight of 390,000,consists of a dimer of 7s yA plus the secretory “piece” (SP) of molecular weight 58,000. Approximately 20 per cent of the SP is bound to the yA portion of the molecule by noncovalent forces and can be released by treatment with agents such as urea or guanidine. The From the Rheumatic Disease Unit, Department of Medicine, State University of New York at Buffalo, Buffalo,New York. THOMAS B. TOMASI, JR., M.D., Ph.D.: Professor of Medicine, Rheumatic Disease Unit, De- partment of Medicine, State University of N e w York at Buffalo, Buffalo, New York 14214. Reprint requests should be addressed to Dr. Tomosi. NUMBER ~~ ARTHRITISAND RHEUMATISM,VOL. 12, No. 1 (FEBRUARY 1969) ~~~~ 45 46 remaining 80 per cent is bound firmly by disulfide linkages and requires reductive cleavage followed by treatment with hydrophobic bond-breaking agents. Whether the noncovalently linked SP is present in those y A molecules ( y A 2 class) which also have noncovalently bonded L chains, as shown by Grey et al.,5 has not as yet been determined. The SP isolated from the intact molecule is identical immunologically to that which occurs free or unattached in the secretions of agammaglobulinemics, newborns, and certain adult secretions such as urine and colostrum. The SP is responsible for the unique chemical and antigenic properties of the 11s secretory molecule, the y A portion of the molecule apparently being identical to serum yA, at least by antigenic analysis. The significance of the SP has not been clearly established although it appears to stabilize the yA molecule, rendering it more resistant to proteolysis with a variety of enzymes. There is no evidence for or against its involvement in the transport of y A from serum to secretions. Moreover, that SP binds nonspecifically to y A and has little or no biological significance has not yet been completely excluded. The information presently available regarding the sites of synthesis of the secretory immunoglobulins is as follows: 1. There is no direct correlation between the serum and salivary levels of yA in diseases associated with either increased or decreased levels of yA in the serum. For example, in yA-type myeloma the salivary levels are essentially norma12 and conversely several reports have appeared of patients with a normal secretory system who apparently completely lack serum yA.6 2. There is a dissociation between the serum and secretory levels in yA during development following birth.? The secretory system appears to mature more rapidly and reaches adult levels of yA much more rapidly than does the serum. yA can be detected in external secretions as early as 10 days of age, frequently at a time when serum yA is demonstrated to be absent by sensitive technics.8 THOMAS B. TOMASI, JR. 3. Radio-labeled serum 7s yA given intravenously to normal individuals does not enter into the saliva.2 Likewise, experiments with exchange transfusions in newborns fail to show significant transport of yA or other immunoglobu1ins.g However, one report has appeared indicating that selective transport of yA into saliva may occur in agammaglobulinemic subjects when high serum levels are obtained by whole plasma infusions.10 In this situation the transport system is specific for yA since yG, which is present in much higher serum concentrations following infusion, does not appear in the parotid fluid. 4. In tissue culture experiments, human parotid and mammary glands incorporate 1%-labeled amino acids specifically into yA.1 It is not known in these studies whether the label is incorporated in the yA portion of the molecule, SP, or both. However, experiments with rabbit mammary gland have suggested that the incorporation is primarily into the SP portion of the secretory molecule.12 5. Fluorescent antibody studies in humans,2 and also in rabbits,l3 have shown an accumulation of immunoglobulin containing plasma cells in the lamina propria of the GI and respiratory tracts and interstitially between the salivary gland acini. yG and yM cells are present in varying proportions but most tissues show a striking predominance of the ybcontaining cells. Using SP specific antisera no staining is seen in the interstitial or lamina propria area and specific fluorescence is localized to the epithelial cells.2 6. In patients with agammaglobulinemia and dysgammaglobulinemia involving a deficiency of serum yA, SP is synthesized in approximately normal amounts.10 Moreover, the epithelial cells of these patients stain with fluorescent SP antisera. The above observations are most consistent with the following interpretation: In the human salivary gland and probably the GI tract and respiratory tract, a significant fraction of the immunoglobulins present in these secretions is synthesized locally in lymphoid cells which are found in close anatomical relationship to the glandular and mucous membrane epithelia. The possibility of some transport from serum cannot be excluded particularly if inflammation of the mucous membrane occurs. The locally produced y A is probably structurally identical with serum y A although this has not been proved. yA traverses the basement ANTIBODIES I N EXTERNAL SECRETIONS membrane of the mucous membrane or glandular acini and then by some as yet unknown mechanism crosses the epithelial membrane. Recent evidence obtained in our laboratoryl4 suggests that transport is in large part through intercellular channels. In many epithelial tissues, such as those in the GI tract and salivary gland, the apical portion of the cell membranes of two adjacent cells are in direct apposition, forming a tight junction15 which presumably would inhibit direct access via the intercellular channels of the secretory macromolecule to the lumen. Because of the finding of a chain determinants along with SP in the apical portion of the epithelial cells, it is hypothesized that the secretory molecule leaves the intercellular space at its apical limit and migrates through the cytoplasm and into the lumen perhaps by a process of reversed pinocytosis. The SPYa nonimmunoglobulin glycoprotein which is intimately associated with cell mucins and which has apparent specificity for the yA molecule, is synthesized by the epithelial cell. It complexes firmly with the (Y chain of yA by both disulfide and noncovalent forces to form the intact 11s secretory molecule. Whether the yA is produced by the plasma cell as an 11s molecule or dimerizes on complexing with SP has not been determined. Also, where the SP and the yA portions unite, whether inside the epithelial cells, on one of its surfaces, or in the lumen, is unknown. The hypothesis presented above concerning the morphological aspects of transport is at present highly speculative and requires further experimental verification. Although, as indicated, it seems likely that in man the majority of yA in most normal external secretions is synthesized locally, there is evidence for individual variation between secretions and in different species. For example, in the case of the GI tract and nasal fluid, a small fraction of yA may 47 be transported from serumlo and this may increase if idammation is present. In the mammary gland of certain species, such as the cow and sheep, there is good evidence that the secretory immunoglobulins are derived from serum and that transport of immunoglobulins is highly se1ective.l' Whether the human mammary gland also shows selective transport rather than local synthesis has not as yet been determined. Thus, although there appears to be species variation and also differences between organs of the same species in regard to sites of synthesis and mechanisms of secretion of immunoglobulins, the common characteristic in all species examined so far is the presence of a specific immunoglobulin class in proportions quite different from those found in serum. This, together with the apparent independent regulation of serum and secretory antibodies in certain situations, appears to justify the separation of the secretory system from that responsible for the production of circulating antibody. The biological significance of the secretory system in terms of its role in normal body defense mechanisms and in various diseases is at present an area of active investigation in many laboratories. The secretions of normal subjects have been shown to contain a number of "natural" antibodies, and many of these have been identified as secretory (11s YA) in type.lS It seems likely that these immune globulins play an important part in the regulation of normal flora and the resistance of mucous surfaces to colonization by potentially pathogenic microorganisms and viruses, although this has not been proved. Moreover, the mechanism by which yA-type antibodies exert their protective action remains unknown, since yA, both that isolated from serum and that in secretions, does not fix complement. It is generally thought that antibodies inhibit the growth of bacteria only 48 THOMAS B. TOMASI, JR. in the presence of complement causing termine whether local immunization is inlysis, or phagocytic cells leading to inges- deed the preferred route. Of considerable interest is the possible tion and killing by intracellular enzyme systems. It has been suggested by Adinolfi participation of the secretory system in et al.1° that secretory yA antibodies are able local allergic reactions of mucous memto lyse Escherichiu coli in the presence of branes. Evidence is available25 that indicomplement and lysozyme, all of these viduals allergic to ragweed have a secretory components being required for lytic action. yA antibody in their nasal secretions which However, these findings are yet to be veri- is capable of interacting specifically with fied in other laboratories. ragweed pollen antigens. Whether it is this Evidence has been presented that re- antibody that is responsible for the clinical covery from certain viral respiratory in- manifestations of ragweed allergy has not fections is better correlated with the titers been shown. Likewise in the GI tract, parof viral-neutralizing antibodies in nasal ticularly in milk and food allergies, it is secretions than with serum titemZ0Similar possible that locally formed secretory anticonclusions have been reached in regard to body is involved in these reactions. respiratory infections with Francisella The role of the secretory system in other tularen.sis,21 in which the serum antibody human diseases has not been clearly deis primarily yM and the nasal secretion fined. It is possible, for example, that local antibody entirely yA. In the GI tract it has antigen-antibody reactions involving the long been known that recovery from ex- secretory system may participate in certain perimental cholera is related to titers of of the so-called autoimmune diseases such coproantibody rather than those in serum.l as Sjogrens syndrome, ulcerative colitis, These observations raise important ques- and pernicious anemia. In this regard it is tions regarding the possible efficacy of local of interest that secretory yA antibody capaimmunization against infectious diseases of ble of binding intrinsic factor has recently the respiratory and gastrointestinal tract. been described in the gastric juice of a paIndeed initial studies have suggested that tient with pernicious anemia.26 aerosol immunization is more effective in In clinical syndromes characterized by a preventing influenzaz2 and t ~ l e r e m i a ~deficiency ~ of serum yA secretory yA is also than is the parenteral route. Oral immuni- absent, although not infrequently replaced zation may also be effective in preventing by other immunoglobulins ( particularly diseases in which colonization occurs first yM) if these are present in the serum. It in the gastrointestinal tract. This is par- seems likely that the deficiency of secretory ticularly well demonstrated by the recent immunoglobulins participates in the lowstudies of Ogra et aLZ4showing the pres- ered resistance to infections which is often ence of yA-type secretory antibody in GI manifested by patients with dysprosecretions following oral polio vaccination teinemias, although this has not been diwhich could be correlated with the ability rectly demonstrated. to prevent the carrier state. Systemic imIn addition to its practical importance in munization (Salk vaccine), although pro- human defense mechanisms and potential ducing similar serum titers, did not result in in immunization programs, the presence of coproantibody and was significantly less the secretory system raises many fundaeffective in preventing the carrier state. mental biological questions for future inFurther field trials are necessary in each of vestigation. How, for example, do the the diseases discussed above in order to de- plasma cells which are in intimate associa- 49 ANTIBODIES I N EXTERNAL SECRETIONS tion with the epithelial tissue differentiate primarily into yA-producing cells? Do they arise, as has been postulated for other immunocompetent cells, from bone marrow precursors and colonize mucous membranes such as those of the respiratory and GI tracts? If this is so, do they arrive in the GI and respiratory tracts precommitted to synthesize yA or is there some influence of the epithelial cells on the differentiation of the secretory lymphoid tissue? What influence, if any, does the thymus have on the differentiation of these cells? Alternatively, is it possible that the secretory lymphoid cells differentiate directly from epithelial tissue as has been suggested for the bursa of Fabricius of chicken^?^' Once SUMMARIO IN the yA is synthesized, what is its route of transport through the epithelium and, since evidence is available that only yA is secreted, what are the molecular receptor mechanisms which determine this specificity? Finally, although considerable emphasis has been placed in this discussion on yA, primarily because of its high concentrations in external secretions and the present lack of information concerning the other immunoglobulins, it should be pointed out that the role of other immunoglobulins and immunoglobulin fragments which are present in these fluids in small amounts may be of considerable importance and deserves more attention in future studies. “TERLINGUA Evidentia de discoperta recente suggestiona que le anticorpore presente in certe secretiones externe es producite localmente e que le imrnunoglobulina predominante in tal secretiones es YA le qua1 es complexe a un glycoproteina nonimmunoglobulinic designate como “pecia” secretori o catena T. Le systema de secretion immunoglobulinic es possibilemente de importantia in determinar le flora local e le resistentia de membranas mucose contra colonisation per potentialmente pathogene bacterios e virus e ha possibilemente u n rolo in reactiones de hypersensibilitate a1 superficies mucose e etiam in certe processos del typo appellate autoimmun. REFERENCES F. A., Hong, R., and Good, R. A.: J. Exp. Med. 1. Burrows, W., and Havens, I.: J. Infect. Dis. 82:231, 1948. 2. Tomasi, T. B., Tan, E. M., Solomon, A., and Prendergast, R. A.: J. Exp. Med. 121:101, 1965. 3. Cebra, J. J., and Robbins, J. B.: J. Immun. 97:12, 1966. 4. Tomasi, T. B., and Czenvinski, D.: In: Bergsma, D. (Ed. ) : Immunologic Deficiency Diseases in Man. Birth Defects Original Article Series, Vol. 4, No. 1, 1968, p. 270. 5. Grey, H., Abel, P., and Kunkle, H.: Fed. Proc. 27:617, 1968. 6. Swanson, V., Dyce, B., Citron, P., Rowleau, C., Feinstein, D., and Haverback, B. J.: Clin. Res. 16:119, 1968. 7. South, M. A., Cooper, M. D., Wollheim, F. A., Hong, R., and Good, R. A,: J. Pediat. 71: 645, 1967. 8. Sullivan, A,, and Tomasi, T. B.: Unpublished observations. 9. Claman, H. N., Merrill, D. A., and Hartley, T. F.: J. Allerg. 40:151, 1967. 10. South, M. A., Cooper, M. D., Wollheim, 123:615, 1966. 11. Hochwald, G. M., Jacobson, E. B., and Thorbeck, G. J.: Fed Proc. 23:557, 1964. 12. Asofsky, R., and Small, P. A.: Science 158: 932, 1967. 13. Crandall, R. B., Cebra, J. J., and Crandall, C. A.: Immunology 12:147, 1967. 14. Tourville, D., Bienenstock, J., Adler, R., and Tomasi, T. B.: J. Exp. Med., Feb., 1969. In press. 15. Farquhar, M., and Palade, G.: J. Cell Biol. 17:375, 1963. 16. Rossen, R. D., Butler, W. T., Vannier, W. E., Douglas, R. G., and Steinberg, A. G.: J. Immunol. 97:925, 1966. 17. Sullivan, A. L., and Tomasi, T. B.: Clin. Res. 12:452, 1964. 18. Tomasi, T. B., and Bienenstock, J.: Advances Immun. 9:1, 1968. 19. Adinol6, M., Glynn, A. A., Lindsay, M., and Milne, C. M.: Immunology 10:517, 1966. 20. Smith, C. B., Purcell, R. H., Bellanti, J. A., THOMAS B. TOMASI, JR. and Chanock, R. M.: New Eng. J. Med. 275:1145, 1966. 21. Bellanti, J. A., Buescher, E. L., Brandt, W. E., Dangerfield, H. G., and Crozier, D.: J. Immunol. 98:171, 1967. 22. Smorodinstev, A. A., and Chalkina, 0. M.: In: Smorodinstev, A. A. (Ed.) : Problems of Pathogenesis and Immunology of Virus Infections. Leningrad, Medgiz, 1955, p. 329. In Russian. 23. Hornick, R. B., and Eigelsbach, H. T.: Bact. Rev. 30:532, 1966. 24. Ogra, P. L., Karzon, D. T., Righthand, F., and MacGillivray, M.: New Eng. J. Med., 1968. I n press. 25. Arbesman, C. E., Dolovich, J., Wicher, K., Dushenski, L. A., Reisman, R. E., and Tomasi, T. B.: Proc. 6th Int. Congr. Aller., Montreal, 1968. In press. 26. Goldberg, L. S., Shuster, J., Stuckey, M., and Fudenberg, H. H.: Science 160:1240, 1968. 27. Ackerman, G. A., and Knouff, R. A.: In: Good, R. A., and Gabrielson, A. E. (Eds.): Thymus in Immunobiology. New York, Hoeber, 1964, p. 123.