Secretory Component in Human Ocular Tissues

SECRETORY COMPONENT IN HUMAN OCULAR TISSUES MATHEA

R. ALLANSMITH,

M.D., AND THOMAS

E. GILLETTE, M.D.

Boston, Massachusetts

Secretory IgA (s-IgA), the predominant class of immunoglobulin in secretory tissues and external secretions 1 such as human tears," is composed of two 7S IgA molecules, one molecule of J chain, and one molecule of secretory component. Secretory component and s-IgA were identified in tears by Josephson and Weiner" and secretory component was identified in human lacrimal tissue by Franklin, Kenyon, and Tomasi.s Using immunofluorescence techniques, we studied the cellular sites for the synthesis of secretory component in lacrimal glands, accessory lacrimal glands, and other ocular tissues. Based on these studies, we designed a model of the ocular secretory system; this model involves: (1) the synthesis of secretory component in lacrimal tissues (both main and accessory glands); (2) the absence of secretory component production in any other ocular tissues, including conjunctiva; and (3) the possible combination of dimeric IgA with secretory component within lacrimal tubules. SUBJECTS

Immunohistochemical studies were performed on exenteration specimens obtained at autopsy from seven subjects with no known ocular disease (Table), on surgical biopsy specimens of lacrimal gland obtained at orbital surgery in three From the Department of Ophthalmology, Harvard Medical School; the Department of Cornea Research, Eye Research Institute of Retina Foundation; and Beth Israel Hospital, Boston, Massachusetts. This study was supported by grant EY-02882 from the National Eye Institute, National Institutes of Health (Dr. Allansmith). Reprint requests to Mathea R. Allansmith, M.D., 20 Staniford St., Boston, MA 02114.

patients (Table), and on biopsy specimens of upper, lower, and bulbar conjunctiva obtained from 14 subjects. The conjunctival specimens were obtained from three patients with no ocular inflammatory disease, from seven with a variety of diseases including three with pemphigoid and one each with quiescent herpetic keratitis, atopic conjunctivitis, giant papillary conjunctivitis, and episcleritis, and f.rom four autopsy cases. For the exenteration specimens the globes were enucleated with a peritomy close to the corneoscleral limbus. The bulbar and palpebral conjunctiva along with the main and accessory lacrimal tissues were then excised en bloc from the orbit (Fig. 1). The postmortem period before exenteration and the clinical status of lacrimal tissue donors are shown in the Table. Tissue specimens were fixed in 19 parts absolute alcohol to one part glacial acetic acid at room temperature for 48 hours, a modification of a technique described by Wolman and Behar.P The tissue was then dehydrated, embedded in degassed paraffin, sectioned at 4 Il. and deparaffinated as described by Sainte-Marie." Tetramethyl rhodamine isothiocyanate (TRITC)-labelled goat antihuman secretory component and fluorescein isothiocyanate (FITC)-labelled rabbit antihuman secretory component were prepared as previously described." The following reagents were obtained commercially: rabbit antihuman IgA and rabbit antihuman IgG (Cappel), rabbit antihuman lactoferrin and rabbit antihuman secretory component (Behring), and FITC-Iabelled goat antirabbit gamma globulin. Purity of antisera was tested by Ouchterlony analysis (Fig. 2).

AMERICAN JOURNAL OF OPHTHALMOLOGY 89:35:h36I , 1980

353

354

AMERICAN JOURNAL OF OPHTHALMOLOGY

MARCH,1980

TABLE LACRIMAL TISSUE DONORS

Case No.

Age (yrs)

Tissue from autopsy

No. of Hours After Death

1

54

22

2

67

23

3

72

20

4

73

22

5

59

120

6

65

9

7

65

31

Tissue from surgery

8

9 10

35 30 31

Diagnosis Alcoholic cirrhosis of liver with jaundice Severe rheumatoid arthritis; pneumonia; subacute bacterial endocarditis Carcinoma pf pancreas with metastasis Progressive supranuclear palsy Cirrhosis; portal hypertension; no jaundice Acute renal tubular necrosis with renal failure; Stapholococcus bacteremia Pulmonary edema; second-degree cardiac failure Thyroid exophthalmos Epibulbar venous aneurysm Thyroid exophthalmos

Tissue sections were stained with the three antisecretory component antisera. The TRITC-conjugated goat antisecretory component was used for direct immunofluorescence staining. Because FITClabelled rabbit antisecretory component produced weak fluorescence, a second layer of FITC-Iabelled goat antirabbit

gamma globulin was added. The unlabelled rabbit antisecretory component was also followed by FITC-Iabelled goat antirabbit gamma globulin. Sections were stained with each antiserum for 15 minutes at room temperature in a moist chamber, then rinsed twice in phosphatebuffered saline, pH 7.2, for five minutes.

Fig. 1 (Allansmith and Gillette). Exenteration specimen. Note inferior fomiceal conjunctiva (black arrow), one of many accessory lacrimal glands (short white arrow), and main lacrimal gland (long white arrow).

Fig. 2 (Allansmith and Gillette). Ouchterlony analysis of reagents. (A) Rabbit antilactoferrin; (B) rabbit anti-IgA; (C) rabbit antilactoferrin); (D) FITC-labelled rabbit anti secretory component. Human tears are in center well. Note lack of identity in lines from wells C and D showing that no antilactoferrin antibodies are in the rabbit antisecretory component reagent.

Fig. 3 (Allansmith and Gillette). Lacrimal glands stained with TRlTC·labelled goat anti-secretory component. Orange indicates specific staining. Blue and green are auto-fluorescence. (A) Main lacrimal gland showing secretory component (orange material) in acinar lumina (long white arrows). Acini devoid of staining for secretory component are also shown (short arrow); bar gauge = 100 /-Lm (x250). (B) Medium-size excretory tubules containing secretory component; bar gauge = 10 /-Lm (x250). (C) High-power view of acini showing diffuse staining of cytoplasm (long white arrow and rimming of lumen (short white arrow); bar gauge = 50 /-Lm (xI,OOO). (0) Accessory lacrimal gland. Note secretory component in acini and lumina; bar gauge = 100 /-Lm (xlOO).

Fig. 4 (Allansmith and GiUette). Main lacrimal gland stained with rabbit anti-secretory component followed by FITC-Iabelled goat anti- rabbit (gG. Compare similarity of staining pattern with Figure 3. Green indicates stain for secretory component. Blue is autofluorescence. (a) Acinus with brightly staining rim of secretory component (long white arrow) leading into tubule with brightly staining cells (short arrows). Note dill'use staining for secretory component C (arrowhead); bar gauge = 100 /-Lm (x250). (b) Small excretory tubule filled with secretory component (short white arrow) surrounded by brightly staining epithelial cells. Dill'usely staining acinus without rlm-statning of luminal surface is shown (long white arrow). Acinus with brightly staining luminar rim and little cytoplasmic staining is shown with arrowhead; bar gauge = 50 /-Lm (x250). (c) Acinus with brightly staining cell (short arrow) adjacent to cell devoid of staining (long arrow); bar gauge = 50 /-Lm •

VOL. 89, NO.3

SECRETORY COMPONENT IN OCULAR TISSUES

Sections were mounted in buffered glycerol, pH 8.0, covered with a cover slip, and sealed with clear nail polish. Control sections for TRITC-stained material were incubated with TRITCconjugated antiserum to bovine serum albumin. Controls of sections for the indirect immunofluorescence technique using FITC label were incubated with normal rabbit serum followed by FITClabelled goat antirabbit gamma globulin. An additional control for secretory component staining was the blocking of fluorescence by incubation of the tissues with unconjugated rabbit antisecretory component followed by rhodamine-conjugated goat antisecretory component. Sections were examined with a Zeiss fluorescence microscope with 2QO-W high-pressure mercury vapor lamp, DC 5 exciter filter, and a 410-nm barrier filter for FITC and a 580-nm excitation filter with a 650-nm barrier for TRITC. Photographs for both fluorescein and rhodamine fluorescence were taken; fluorescein with the observation filter combination, rhodamine with a BC12 exciter filter, and 530-nm barrier filter combination. Ektachrome ASA 200 and ASA 400 color films were used.

357

were seen in which brightly staining cells were adjacent to negative cells (Fig. 4). Entire acini were seen that were completely devoid of staining (Fig. 3). Acinar lumina often contained brightly staining material and the luminal rim exhibited a strong fluorescence (Fig. 3). This phenomenon was seen both in acini containing secretory component and in acini devoid of intracellular staining. No intercellular staining was detected. The excretory tubules were identified by their larger luminal diameter relative to the epithelial circumference. The tubules ranged from those contiguous with acini and of one epithelial layer (Fig. 4) to large tubules distant from acini and of two or three layers of epithelium with no adjacent interstitial plasma cells (Fig. 5). In contrast to acinar cells, most staining

RESULTS

Antisecretory component antisera stained the secretory and tubular areas of all lacrimal tissues, but the pattern of staining differed for the two areas (Figs. 3 and 4). Staining of acini, identified by their small luminal diameter relative to epithelial circumference, was seen in approximately 60% of acinar cells. The intensity of staining ranged from barely perceptible (most cells) (Fig. 3) to extremely bright (about 5%). The differences in the intensity of staining may be attributed to a maturational difference rather than to a difference in cell type. Staining was seen both in basal and apical portions of the acinar cells, and acini

Fig. 5 (Allansmith and Gillette). Large excretory tubules of main lacrimal gland stained for secretory component. Note bright staining of many tubular cells (arrows), absence of plasma cells in interstitial areas around tubules, and nonstaining acini (arrowheads). Bright material in vessel (V) is autofl.uorescent; bar gauge-SO ....m (x2SO).

358

AMERICAN JOURNAL OF OPHTHALMOLOGY

tubular cells were brightly stained. Tubules were seen in which all epithelial cells stained brightly; this phenomenon was not seen in acini. Bright staining was seen in tubules of all sizes, ranging from those with a single layer of cuboidal epithelium (Fig. 3) to those composed of two or three layers of epithelial cells (Fig. 5). The staining pattern of some tubules was similar to that of some acini, in that cells without stain were adjacent to staining cells (Fig. 4). In general, approximately 30% of tubular cells stained brightly compared with approximately 5% of acinar cells. Some tubules were sectioned tangentially and showed positive staining of the epithelium throughout their length against a background of lesser-staining acini (Fig. 4). Tubular lumina were often filled with intensely staining material (Figs. 3-5). This was usually found in conjunction with positively stained epithelial cells, although it was also seen in the presence of unstained tubular epithelial cells. No staining of interstitial tissues, vascular components, or lymphoid elements was seen, with the exception of interstitial staining in Case 1. Accessory lacrimal glands exhibited a secretory component staining pattern identical to that of the main lacrimal glands (Fig. 3). Accessory glands were available only from autopsy cases. All sections of conjunctiva examined, whether from biopsy or autopsy specimens, revealed no staining for secretory component within the epithelium or substantia propria (Fig. 6). Staining of a homogeneous material adherent to the conjunctival surface was seen in some specimens (Fig. 6, inset). No staining for secretory component was observed in the cornea, ms, choroid, retina, sclera, episclera, extraocular muscles, or orbital fat. A 54-year-old man had alcoholic cirrhosis (Case 1, Table) and was deeply jaun-

MARCH, 1980

Fig. 6 (Allansmith and Gillette). Conjunctiva stained for secretory component. Note absence of specific staining in epithelium and substantia propria. Staining was occasionally seen on the surface that corresponded to mucus layer (arrow); bar gauge-50 urn (x250).

diced at the time of death. The interstitial tissue stained moderately for secretory component with all three antisera. The stained material was diffuse and differed notably from tissues of the other nine subjects. DISCUSSION

Areas of the normal human eye that are associated with large accumulations of plasma cells adjacent to epithelial mucosal surfaces include conjunctiva, lacrimal gland, and accessory lacrimal glands." Those areas that contain secretory component are also potential sites for synthesis of s-IgA. Our results confirm those of Franklin and coworkers"; secretory component is produced in the human lacrimal gland. This study extends their work in describing that over 60% of the acini stain for secretory component. With stronger antisera, staining might be even greater. In about 5% of the cells, secretory component within the acinus was at such a level as to stain brightly. We add the previously unreported finding that tubules stain brightly for secretory component, which indicates that they might be a site for the production of secretory component, and that accessory lacrimal tissue forms secretory component. Possibly, the large amount of secretory

VOL. 89, NO.3

SECRETORY COMPONENT IN OCULAR TISSUES

359

Fig. 7 (Allansmith and Gillette). Diagrammatic representation of secretory component formation in lacrimal glands. Dirneric IgA coupled with J chain (A-J-A) is Formed by plasma cells in interstitial tissues of main and accessory lacrimal glands. This molecule is combined with acinar-produced secretory component to make s-IgA, which is excreted into lumen, A small amount of dimeric IgA free of secretory component is excreted into lumen. Free secretory component is also excreted by tubules into lumen to create relative excess of secretory component that combines with luminal dimeric IgA; s-IgA flows onto all areas of ocular surface.

component found in tubular cells combines with free dimeric IgA intraluminally. A similar distribution of cells that secrete a high level of secretory component in the tubules as compared with the acini was noted also in the mouse submaxillary gland." Intraluminal combina-

tion of dimeric IgA with secretory component was also suggested for this system. A schematic representation of our concept of the ocular s-IgA system is presented in Figure 7. Excretory tubules contain secretory component (Fig. 5) but lack secretory

360

AMERICAN JOURNAL OF OPHTHALMOLOGY

granules.!'' which precludes the presence of secretory component in these granules. Whereas Franklin and co-workers-! reported the presence of secretory component in the rabbit conjunctiva, we were unable to detect secretory component in human conjunctiva. This discrepancy may result from an interspecies difference. Franklin (personal communication, 1978) reported that human autopsy material stained irregularly, whereas surgical material showed secretory component more reliably. The report of Franklin and associates was based on lacrimal specimens from surgery. Although most of our tissues were obtained from autopsies (Table), the staining patterns were identical for tissues from the two sources. The autopsy specimens were preferable because they allowed examination of more areas. A crucial aspect of the localization of secretory component is the antisera used. Lactoferrin is difficult to separate from secretory component, and antilactoferrin antibodies therefore often contaminate secretory component antisera. 1 The lack of antilactoferrin antibody in our preparations was shown for each antiserum by Ouchterlony analysis (one example shown in Figure 2). The high level of secretory component observed within the ducts might have been caused by the absorption of secretory component from the lumen into the cells lining the tubules. This possibility was excluded as we observed some tubules, large and small, that had highly staining luminal material within but no staining of adjacent epithelial cells (Fig. 3). Case 1 had alcoholic cirrhosis with jaundice, and secretory component has been recorded to increase an average of sixfold in the serum of persons with cirrhosis.P This patient exhibited secretory component in interstitial spaces of Iacri-

MARCH, 1980

mal glands. Possibly, in Case 1 (Table), secretory component may have been present in the serum at a level sufficient to cause interstitial staining. We conclude that the sites of secretory component manufacture around the eye are the acini of the main and accessory lacrimal glands and the excretory tubules of lacrimal tissue. We also conclude that, in the human, the conjunctiva and other ocular and periocular tissues do not stain for secretory component by the techniques used. SUMMARY

Lacrimal tissue and accessory lacrimal tissue from seven autopsy cases, lacrimal tissue from three patients undergoing orbital exploration, and biopsy specimens from conjunctiva of 14 subjects were examined for the presence and distribution of secretory component by three antisecretory-component antisera. Secretory component was present in all lacrimal and accessory lacrimal tissues but in no other ocular tissues. Over 60% of acinar cells stained for secretory component; about 5% of acinar cells stained brightly; about 30% of tubular cells stained brightly. Main and accessory lacrimal tissues appeared identical in their staining patterns. We concluded that the main sites of synthesis of secretory IgA in human ocular tissues are the lacrimal and accessory lacrimal tissues. ACKNOWLEDGME:"TS

Richard Dallow, M.D., provided surgical specimens and Sylvia Crago, M.D., and Jiri Mestecky M.D., provided antisera. REFERENCES 1. Tomasi, T. B.: The Immune System of Secretions. Englewood Cliffs, New Jersey, Prentice-Hall Foundations of Immunology Series, 1976, p. 143. 2. Chodirker, W., and Tomasi, T. B., jr.: Gamma globulins. Quantitative relationships in human serum and non-vascular fluids. Science 142: 1080, 1963. 3. Josephson, A. S., and Weiner, R.: Studies of the

VOL. 89, NO.3

SECRETORY COMPONENT IN OCULAR TISSUES

proteins of lacrimal secretions. J. Immuno!' 100: lOBO, 1968. 4. Franklin, R. M., Kenyon, K. R., and Tomasi, T. B., jr.: Immunohistologic studies of human lacrimal gland. Localization of immunoglobulins, secretory component and lactoferrin. J. Immunol. 110:984, 1973. 5. Wolman, M., and Behar, A.: A method of fixation for enzyme-cytochemistry and cytology. Exp. Cell Res. 3:619, 1952. 6. Sainte-Marie, G.: A paraffin embedding technique for studies employing immunofluorescence. J. Histochem. Cytochem. 10:250, 1962. 7. Crago, S. S., and Mestecky, J.: Secretory component. Interactions with intracellular and surface immunoglobulins of human lymphoid cells. J. Immuno!' In press. 8. Allansmith, M. R., Kajiyama, G., Abelson, M.

361

B., and Simon, M. A.: Plasma cell content of main and accessory lacrimal glands and conjunctiva. Am. J. Ophthalmol, 82:819, 1976. 9. Comoglio, P. M., and Guglielmone, R.: Immunohistochemical study of IgA transepithelial transfer into digestive tract secretions in the mouse. Immunology 25:71, 1973. 10. Orzalesi, N., Riva, A., and Testa, F.: Fine structure of human lacrimal gland. 1. The normal gland. J. Submicr. Cytol. 3:283, 1971. II. Franklin, R. M., Prendergast, R. A., and Silverstein, A. M.: Secretory immune system of rabbit ocular adnexa, abstract. Presented before the Association for Research in Vision and Ophthalmology meeting, Sarasota, Florida, May 2, 1978. 12. Andre, F., and Andree, C.: Cirrhotic glomerulonephritis and secretory immunoglobulin A. Lancet 1:197, 1976.

Fifty years ago this month in The [oumal: In the whole field of ophthalmic surgery, no maneuver calls for greater delicacy of touch than this grasping of a delicate anterior capsule with the intent to employ that capsule as a means of traction upon the suspensory ligament. Failure, in properly selected cases, may depend first upon attempting too large a grasp upon the anterior capsule, second upon holding the fold of capsule too firmly, and third upon using traction too brusquely. Intracapsular extraction, By W. H. Crisp Am. ]. Ophthalmol. 13:244, 1930