- Email: [email protected]
Immune capacity of the chicken bursectomized at 60 Hr of incubation: Cytoplasmic immunoglobulins and histological findings
CLINICAL
IMMUNOLOGY
AND
IMMUNOPATHOLOGY
30, 41-50 (1984)
Immune Capacity of the Chicken Bursectomized Incubation: Cytoplasmic lmmunoglobulins Histological Findings’ SIRPA JALKANEN,~ RITVA KORPELA, KAISA PAAVO Departments
of Medical
at 60 Hr of and
GRANFORS,
AND
TOIVANEN
Microbiology SF-20520 Turku
and Pathology, 52, Finland
Turku
University,
Chickens were surgically bursectomized at 60 hr of incubation, before the bursal anlage appears. Completeness of the bursectomy was confirmed at autopsy at 10 weeks of age. These embryonically bursectomized (Bx)~ chickens are known to produce immunoglobulins of IgM, IgG, and IgA classes but so far no specific antibodies have been observed even after heavy immunization. The Bx chickens had mature plasma cells in an almost normal frequency when studied at 10 weeks of age. The amount of germinal center formation in the spleen and cecal tonsils was markedly decreased when compared to the control (Co) chickens. Also, the frequency of cytoplasmic h&positive (c-IgA+) cells was severely decreased in the Bx animals, whereas the occurrence of c-IgG+ and c-IgM+ cells was not affected to the same extent. These findings support the hypothesis that heavy-chain class switch may occur without the bursal influence, and that the bursa of Fabricius is essential only for expansion or creation of the antibody repertoire.
INTRODUCTION
Different opinions have been presented about the significance of the avian bursa of Fabricius in B-cell differentiation (1). Grossi et al. (2) detected B lymphocytes expressing simultaneously surface and cytoplasmic IgM in the bursa of Fabricius from IZday embryos lacking detectable Ig-positive cells elsewhere. Both the size of the B-lymphocyte pool and the generation of the precursors secreting different immunoglobulin classes are controlled by the bursa (3). Also, the normal switch from IgM to IgG production is thought to take place in the bursa (4). On the other hand, previous studies by us (5-7) and others (4, 8-14) provide strong evidence for the existence of extrabursal sites for B-cell development. We have found that chickens surgically bursectomized at 60 hr of incubation produce serum and surface immunoglobulins of IgM, IgG, and IgA classes, but are not able to form specific antibodies. These chickens closely resemble human patients suffering from antibody deficiency with a normal level of serum immunoglobulins (15, 16). Fitzsimmons et al. (10) reported cortical atrophy of the thymus after early I This work was supported by the Research and Science Foundation of L%ke Oy, The Emil Aaltonen Foundation, and The Finnish Cancer Union. 2 To whom correspondence should be addressed: Department of Medical Microbiology, ‘Ibrku University, SF-20520 ‘Ihrku 52, Finland. 3 Abbreviations used: Bx, bursectomized; Co, control; c-Ig’ , cytoplasmic immunoglobulin positive; LPS, lipopolysaccharide; DNPrBSA, dinitrophenyl-bovine serum albumin; PBS, phosphate-buffered saline; s-Ig+, surface immunoglobulin positive. 41 0090-1229/84 $1.50 Copyright B 1984 by Academic Press, Inc. All rights of reproduction in any form reserved
42
JALKANEN
ET Al
embryonic bursectomy and concluded that normal embryonic development of the thymus requires the presence of the bursa, and perhaps a subpopulation of T cells does not develop at all without the bursal influence. In contrast, JankoviC et al. (12) could not find any changes in the cellular makeup of the thymus following early embryonic bursectomy. Likewise, cell transfer studies have provided no evidence for the existence of bursaderived T cells (17, 18). In this work we report the existence and distribution of cytoplasmic immunoglobulin positive (c-Ig+) cells and histological findings in the spleen, cecal tonsils, Harderian gland, bone marrow, and thymus of the embryonically bursectomized (Bx) and control (Co) chickens at the age of 10 weeks. MATERIALS
AND METHODS
Chickens. Fertilized eggs from the lines P (genotype B2B2) and V (genotype B15Bt5), kept at the Department of Medical Microbiology, Tbrku University, were used. The incubation of the eggs and the care of the birds were as described previously (19). Bursectomy. Bursectomy at 60 hr of incubation was carried out according to the previously described techniques (9, 11) with some modifications. The operation was performed under aseptic conditions using microsurgery scissors just before the amniotic membrane began to cover the caudal part. The tail was cut off caudally to the leg buds. Hatchability of the operated embryos was IO-20% with 2% surviving beyond the first week of life. The controls consist of age and sex-matched normal chickens. Antigens and immunizations. The chickens were immunized at the age of 6 weeks with Escherichia coli lipopolysaccharide (LPS), levan, tetanus toxoid, and dinitrophenyl-bovine serum albumin (DNPrBSA). The origin and amount of the antigens given at a time were as follows: E. coli LPS (lipopolysaccharide W, E. coli 055:B5, Difco Laboratories, Detroit, Mich.) 10 Fg, native levan prepared from Corynebacterium levaniformis as described previously (20,21) 100 kg, commercially licensed tetanus vaccine (The National Public Health Institute, Helsinki, 10 Lf/ml tetanus toxoid) 1 ml of 1:20 dilution, and DNP,BSA, prepared according to the method of Eisen (22), 1 mg. Each antigen was administered intraperitoneally four times with l-week intervals. Autopsies and histology. For the autopsies at 10 weeks of age the chickens were sacrificed by exsanguination. To determine completeness of the bursectomy, the whole cloaca1 region was carefully dissected out, fixed with 10% neutral buffered formalin, and embedded in paraffin. Serial sections cut throughout the region were stained with van Gieson and hematoxylin and eosin. The spleen, bone marrow, one to two thymic lobes, Harderian gland, and cecal tonsils were processed in the same way and stained with van Gieson, hematoxylin-eosin, periodic acid-Schiff, and methyl- green-pyronine stains. Alcian blue mucus staining was used for samples from the Harderian gland and the cecal tonsils. Antisera. Rabbit anti-chicken y (Miles-Yeda, Rehovot, Israel) was further purified by absorption with Sepharose 4B-conjugated chicken IgM and IgA. Goat anti-chicken (Y (Miles-Yeda) and goat anti-chicken p, (Miles-Yeda) used had only
BURSECTOMY
AT 60 HOURS
OF
INCUBATION
43
anti-heavy chain specificity. Specificity of all antisera used in this study was ascertained as described previously (5). Immunoperoxidase procedure. A four-step immunoperoxidase technique was used to stain 4 p,m thick paraffin-embedded tissue sections (23). The sections were deparaffinized with xylene and alcohol and treated with pepsin for 2 hr at 37°C to release bound antigens. Endogenous peroxidase was blocked by 30 min of incubation in methanol containing 1.6% H202, and the sections were flooded with normal swine serum (Dak~immunoglobulins A/S, Copenhagen, Denmark) 1:50 for 10 min to diminish nonspecific background staining. At the first step the sections were incubated with 1:lOOO dilution of rabbit anti-chicken y, goat antichicken lo or 01(Miles-Yeda), or normal rabbit or goat serum as controls at 37°C for 1 hr. After three washings with phosphate-buffered saline (PBS) the sections were incubated with either swine anti-rabbit or anti-goat immunoglobulins (Dako) I:50 at 37°C for 30 min and then with rabbit or goat antiperoxidase-peroxidase complex (Dako) 1:500 at 37°C for 15 min. After three washings with PBS and flooding with fresh diaminobenzidinehydrogen peroxide the sections were again washed with PBS, counterstained with hematoxylin, and mounted for examination by a light microscope. In the quantitative evaluation of cells stained with anti-p, anti-y, or anti-a, the sections were first screened using a x 40 magnification; 10 fields with the highest frequency of c-Ig+ cells were chosen for further enumeration with x400 magnification. The analyses were carried out without knowing to which experimental group the subject belonged. RESULTS
All Bx birds appeared typically tailless. One out of seven Bx chickens died during immunization. The others survived in good condition to the end of the experiment. General Histology Autopsy and microscopical examination revealed no bursal remnants in the Bx chickens. Active germinal centers were observed in the spleen and cecal tonsils both in the Bx and Co groups (Fig. 1). In the cecal tonsils of the Bx chickens the number of germinal centers was only one-tenth of the Co values and in the spleen even less. Nevertheless, the number of mature plasma cells reached roughly the same level in the Bx and Co chickens except for one Bx bird (No. 2290), which had very few mature plasma cells in all the organs studied. The structure of the thymus and Harderian gland of the Bx chickens appeared normal, without any signs of atrophy. No essential difference was observed in the bone marrow sections of the Bx and Co animals. Antibody Response None of the Bx birds showed any evidence for production of specific antibodies against any of the antigens used for immunization, not even after the fourth antigenic injection. The determinations were made using enzyme-linked immu-
44
JALKANEN
ET AL
FIG. I. A germinal center (gc) surrounded by the capsule in the cecal tonsils of a Bx bird. On the left is a gland duct (gd). c-IgG+ cells stained by immunoperoxidase appear both inside and outside the germinal center. ( x 400).
nosorbent assay (5). Detailed results of these as well as production of various natural and autoimmune antibodies will be reported in a subsequent paper of this series. c-Ig + Cells c-&M. All Bx birds had c-IgM+ cells. Regarding their distribution and frequency, two findings deserve attention. (i) One of the six Bx birds (No. 2290) had very few c-IgM+ cells, whereas all the others had them in an approximately normal frequency. (ii) On average, decreased frequency of c-IgM+ cells was observed only in the bone marrow of the Bx birds: 1.9 k 0.7 cells/microscopic field compared to 3.9 k 1.1 in the Co animals (P < 0.005, Table 1). c-ZgG. The findings are very much the same as those described regarding the occurrence of c-IgM + cells. The Bx bird No. 2290 had very few c-IgG + cells, whereas on average the frequency of c-IgG+ cells was decreased only in the Harderian gland (P < 0.025) and in the bone marrow (P < 0.02, Table 2). c-ZgA. c-IgA+ cells are affected by the early bursectomy considerably more clearly than c-IgM + or c-IgG + cells. On average, c-IgA+ cells occurred in a significantly decreased frequency in the Harderian gland, spleen, and particularly in the cecal tonsils of the Bx birds, whereas for the bone marrow and thymus the
BURSECTOMY
45
AT 60 HOURS OF INCUBATION
TABLE 1 c-IgM+ CELLS IN LYMPHOID OUGANS OF INDIVIDUAL Bx CHICKENS AND OF THEIR CONTROLS(Co) AT THE AGE OF 10 WEEKS Bx chickens
Co chickens
Organ
2282
2290
2291
2292
2288
2323
Mean f SD6 Mean + SDb
Harderian gland Cecal tonsil Spleen Bone marrow Thymus
15.5~ 5.1 3.8 2.2 3.2
0.0 0.3 0.0 1.2 0.0
18.0 2.1 1.2 1.9 0.0
10.5 2.0 0.4 1.6 0.6
28.0 6.6 0.6 3.2 0.1
43.0 34.3 0.3 1.4 0.4
29.2 8.6 1.0 1.9 0.7
+ f + + *
27.1 12.8 1.4 0.7c 1.2
53.2 3.6 2.0 3.9 0.5
-t ” + t IT
19.4 1.0 3.5 1.1 0.8
a c-IgM+ cells/field ( x 400; mean of IO fields). “n=6. ’ P < 0.005, when compared to the Co group by Student’s t test,
mean Bx values were not significantly different from the corresponding Co values (Table 3). Generally, the distribution of c-Ig+ cells, both of morphologically mature plasma cells and of lymphoblasts, was similar in different tissues of Bx and Co chickens. Intracellular localization of immunoglobulins was homogeneous or fine granular with no differences between the Bx and Co animals. Both morphologically mature plasma cells and c-Ig+ lymphoblasts were found in all the peripheral lymphoid organs studied and sporadically also in the thymus. In the bone marrow all c-Ig+ cells were lymphoblasts, immature lymphocytes, and lymphocytes of medium and large size (24), both in Bx and Co chickens; no mature plasma cells were found in the bone marrow. In the Harderian gland c-Ig + cells appeared in clusters and were usually stained more intensely than in the other organs. In the spleen c-Ig+ cells were mostly seen in the mantle zones of germinal centers or surrounding them. In the cecal tonsils the c-Ig+ cells were localized in the lamina propria and germinal centers. c-Ig+ plasma cells were also observed in other parts of the intestinal mucosa. TABLE 2 c-IgG+ CELLS IN LYMPHOID ORGANS OF INDIVIDUAL Bx CHICKENS AND OF THEIR CONTROLS(CO) AT THE AGE OF 10 WEEKS Bx chickens Organ Harderian gland Cecal tonsil Spleen Bone marrow Thymus
Co chickens
2282
2290
2291
2292
2288
2323
Mean + SDb
Mean t SDb
50.7” 35.3 0.3 2.7 0.0
0.0 1.8 0.0 2.8 0.0
33.5 4.7 0.0 2.1 0.0
74.5 4.9 0.2 2.9 0.2
97.0 1.0 0.1 1.0 0.0
77.0 24.4 11.3 1.3 0.1
55.5 12.0 1.9 2.1 0.05
95.6 21.1 2.2 4.2 0.5
” c-IgG + cells/field ( x 400; mean of 10 fields). “n=6. c P < 0.025, when compared to the Co group by Student’s t test.
dP < 0.02.
2 ? ” -e 2
35.0c 14.3 4.6 o.a* 0.08
I 2 “_ + ‘-
12.9 6.7 1.8 1.6 0.8
46
JALKANEN
ET AL
TABLE 3 c-IgA+ CELLS IN LYMPHOID
ORGANS
OF INDIVIDUAL Bx CHICKENS THE AGE OF 10 WEEKS
AND OF THEIR CONTROLS
Bx chickens Organ Harderian gland Cecal tonsil Spleen Bone marrow Thymus
2282
2290
2291
2292
2288
2323
8.1” 1.4 0.1 0.4 0.0
0.0 0.0 0.0 0.5 0.1
2.8 0.2 0.0 1.6
7.1 1.0 0.0
5.0 0.9 0.0
1.1 0.0
2.0 0.0
2.9 0.5 0.0 0.3
0.1
0.0
Mean * SD6 4.3 0.7 0.02 I.0 0.03
_t t * 2 +
3.Od 0.5’ 0.04’ 0.7 0.05
(Co)
AT
Co chickens Mean 2 SDb 14.9 12.1 0.8 1.8 0.2
k -r k 5 2
5.1 5.9 0.7 0.8 0.2
a c-IgA+ cells/field (x400; mean of IO fields). bn = 6. c P < 0.025, when compared to the Co group by Student’s t test.
dP < 0.005. PP < 0.001.
In the cecal tonsils and in the Harderian gland apical parts of the epithelium were IgA positive (Fig. 2), and IgM and IgG staining of these parts was negative (Fig. 3). Also the mucus in the lumen was IgA positive, indicating active secretion of IgA.
FIG. 2. c-IgA+ cells in the Harderian gland of the Bx bird stained by immunoperoxidase. A typical plasma cell with IgA-positive cytoplasm is marked with an arrow. The apical part (two arrows) of the epithelium is stained IgA positive also. (x 1600).
BURSECTOMY
AT 60 HOURS OF INCUBATION
47
FIG. 3. c-IgM+ cells in the cecal tonsils of a Bx bird stained by immunoperoxidase. A typical plasma cell with IgM-positive cytoplasm is indicated with an arrow. Apical secretion of IgM is missing. (X looo).
DISCUSSION
The majority of c-Ig+ cells of the Bx birds had a typical plasma cell morphology, but immature lymphocytes and lymphocytes of medium and large sizes also contained c-Ig. In this respect the early bursectomy seems to have had no significant effect. The most striking effect of the bursectomy was related to the occurrence of c-IgA+ cells in the Harderian gland, cecal tonsils, and spleen. In all these organs the Bx birds had significantly fewer c-IgA+ cells than the controls (Table 3). Most probably this applies also for the thymus, where the frequency of cIgA+ cells as such is very low (Table 3). It is difficult to say whether the decrease of c-IgA+ cells in the Harderian gland, cecal tonsils, and spleen is or is not compensated by an increase of c-IgM+ cells, which occur in quite a normal frequency in these organs, but are decreased in the bone marrow (Table 1). Our previous studies have demonstrated decreased frequency of surface IgA positive (s-IgA+) cells (spleen) and of s-IgG+ cells (spleen, thymus, peripheral blood) in the Bx chickens, together with a minimal amount of serum IgG (5, 6). The decreased occurrence of germinal centers in the Bx birds is a remarkable finding. However, it is related to the defect in antibody formation and IgG production found typically in the Bx birds (5-7). Otherwise, no structural changes, hypoplasia, or atrophy were observed in the peripheral lymphoid organs of the
48
JALKANEN
ET AL
Bx birds at 10 weeks of age. Fitzsimmons and co-workers (IO, 25) found marked cortical hypoplasia of the thymus in the chickens bursectomized as early embryos. It is possible that lymphoid structure of the thymus may develop more slowly in Bx than in normal chickens. This can partially explain the discrepancy with our observations, since Fitzsimmons et al. mostly analyzed their birds as embryos. Our findings are in concordance with those of Jankovic et al. (12), who found no changes in the cellular makeup of the thymus in the chickens bursectomized at 52-64 hr of incubation. On this basis, the bursa of Fabricius seems not to have any marked effect on the structure of other lymphoid organs. At the present, a meaningful synthesis of the findings concerning Bx chickens seems difficult. A summary of the occurrence of c-Ig + and s-Ig + cells is presented in Table 4. Data on the thymus have been excluded, due to the low frequency of B cells or their precursors in this organ and to the uncertainties caused by it in the interpretation. Likewise, in the spleen the statistical significance of the decreased frequency of cells with c-IgM, s-IgM, or c-IgG is hampered either by large individual variations or by low values. It is clear, however, that c-Ig’ and s-Ig+ cells of all three isotypes are found in the Bx birds, indicating, as also stated before (5-7), that switch in isotype expression occurs independently of the bursa of Fabricius. It is possible that Bx birds have a disturbance in the regulation of the switch since the occurrence of cellular IgA and IgG seems to be affected by the early bursectomy more than that of cellular IgM, particularly in the peripheral lymphoid tissues. In addition, Bx chickens are known to be unable to produce specific antibodies in spite of immunoglobulin production (5, 6). These findings have led us to suggest that the bursa of Fabricius is not essential for the isotype switch, but rather only for the expansion or creation of the antibody repertoire. The explanation for the occurrence of immunoglobulins without antibody specificity remains to be found. One could argue that the bursectomies were subtotal and a few follicles which were sufficient for isotype switch, but not for the genTABLE FREQUENCY
OF c-Ig+
CELLS BASED
AND
s-Ig+
ON DATA
CELLS
4
IN DIFFERENT
PRESENTED
IN THIS
AND
ORGANS
OF lo-WEEK-OLD
A PREVIOUS
Bx
CHICKENS.
PAPERY
Cells Organ Harderian gland Cecal tonsils Spleen Peripheral blood Bone marrow
c-IgM +
s-&M +
c-&G+
s-IgG +
Normal Normal Normal ND JJ
NDb ND Normal 11 t
i’ Normal Normal ND -1
ND
c-IgA + l’J”le
rNl? JJJ Normal
:i~ Normal
s-IgA + ND ND i-1 Normal Normal
n S. Jalkanen, K. Granfors, M. Jalkanen, and P. Toivanen, J. Immunol. 130, 2038, 1983. b Not determined. c J Slightly decreased ( t slightly increased); P 5 0.05 when compared to normal controls by Student’s t test. d & i Decreased; P 5 0.01. e 4 4 J Markedly decreased; P 5 0.001.
BURSECTOMY
AT 60 HOURS OF INCUBATION
49
eration of antibody diversity, developed. However, this possibility was excluded by careful microscopic examination. All of the chickens studied appeared completely tailless, lacking also parts of the coprodeum, the most cranial part of the avian cloaca. These findings also exclude the possibility that a few bursal follicles developed after the bursectomy and were already in a complete involution at the time of the autopsy. There were birds in which microscopic examination revealed a few bursal follicles. However, all of these birds had a considerable part of the cloaca and a clearly demonstrable antibody response against all the antigens used; these chickens were excluded from the study. In the bone marrow, decrease of c-IgM+ cells and increase of s-&M+ probably argues for a disturbance in the maturation of B cells. In spite of this, bone marrow appears more normal than the other tissues studied regarding the occurrence of c-Ig+ or s-Ig+ cells. Bone marrow cells of Bx birds respond in a normal fashion by cell proliferation to anti-p or anti-y, whereas spleen cells of Bx birds respond significantly less than spleen cells of normal chickens (7). However, it is obvious that more definite explanations for these peculiarities have to wait for molecular analyses in experimental animals or human patients (16). ACKNOWLEDGMENTS We thank Dr. Lauri Pelliniemi for advice in immunoperoxidase methodology, Soile Niittoaho, Tarja Puurunen, Marjo Vesanto, and Anja Vuoristo for expert technical assistance, and Eija Nordlund for secretarial help.
REFERENCES I. Toivanen, A., Toivanen, P., Eskola, J., and Lassila, O., In “Avian Immunology” (M. E. Rose, L. N. Payne, and B. M. Freeman, Eds.), pp. 45-62, British Poultry Science Ltd., Edinburgh, 1981. 2. Grossi, C. E., Lydyard, P. M., and Cooper, M. D., J. Immunol. 119, 749, 1977. 3. Lawton, A. R., Kincade, P. W., and Cooper, M. D., Fed. Proc. 34, 33, 1975. 4. Subba Rao, D. S. V., McDufie, F. C., and Glick, B., J. Immunol. 120, 783, 1978. 5. Granfors, K., Martin, C., Lassila, O., Suvitaival, R., Toivanen, A., and Toivanen, P., Clin. Immunol.
6. 7. 8. 9. 10. 11. 12. 13. 14. IS. 16. 17. 18.
Immunopathol.
23, 459, 1982.
Jalkanen, S., Granfors, K., Jalkanen, M., and Toivanen, I?, J. Immunol., 130, 2038, 1983. Eerola, E., Jalkanen, S., Granfors, K., and Toivanen, A., J. Zmmunol., 131, 120, 1983. Lemer, K. G., Glick, B., and McDufEe, F. C., J. Immunol. 107, 493, 1971. Fitzsimmons, R. C., Garrod, E. M. F., and Garnett, I., Cell. Immunol. 9, 377, 1973. Fitzsimmons, R. C., Dixon, D. K., and Kocal, E. M. F., In “Developmental Immunobiology” (J. B. Solomon and J. D. Horton, Eds.), pp. 387-394, Elsevier/North-Holland, Amsterdam, 1977. Jankovic, B. D., Knezevic, Z., Isakovic, K., Mitrovic, K., Markovic, B. M., and RajCeviC, M., Eur. J. Immunol. 5, 656, 1975. Jankovic, B. D., Isakovic, K., Markovic, B. M., RajCeviC, M., and Knezevic, Z., Exp. Hematol. 4, 246, 1976. Jankovic, B. D., Isakovic, K., Markovic, B. M., and RajCeviC, M., Immunology 32, 689, 1977. Sato, K., and Abe, S., Immunology 28, 293, 1975. Rothbach, C., Nagel, J., Rabin, B., and Fireman, P., J. Pediatr. 94, 250, 1979. Saxon, A., Kobayashi, R. H., Stevens, R. H., Singer, A. D., Stiehm, E. R., and Siegel, S. C., Clin. Immunol. Immunopathol. 17, 235, 1980. Lassila, O., Acta Pathol. Microbial. &and. [C’j 87, 287, 1979. Weber, W. T., and Alexander, J. E., J. Zmmunol. 121, 653, 1978.
50
JALKANEN
ET AL
19. 20. 21. 22. 23.
Toivanen P., and Toivanen, A., Eur. J. Immunol. 3, 58S, 1973. Miranda, J. J., Immunology 23, 829. 1972. Moreno, C., Courtenay, B. M., and Howard. J. G., fmmu~oc~ernisf~ 13, 429, 1976. Eisen, H. N., Belman, S., and Carsten M. E.. J. Amer. Chem. Sot. 75, 4.583. 1953. Sternberger, L. A., Hardy, P. H.. Jr., Cuculis, J. J., and Meyer, H. G., J. Histochem. Cytochem 18, 315, 1970. 24. Lucas, A. M., and Jamroz, C., “Atlas of Avian Hematology,” Agriculture Monograph 25. tJ.S Department of Agriculture, Washington, D.C., 1961. 25. Dixon, D. K.. and Fitzsimmons, R. C., Dev. Comp. Immunol. 4, 713, 1980. Received February 8, 1983; accepted with revision May 31, 1983.
IMMUNOLOGY
AND
IMMUNOPATHOLOGY
30, 41-50 (1984)
Immune Capacity of the Chicken Bursectomized Incubation: Cytoplasmic lmmunoglobulins Histological Findings’ SIRPA JALKANEN,~ RITVA KORPELA, KAISA PAAVO Departments
of Medical
at 60 Hr of and
GRANFORS,
AND
TOIVANEN
Microbiology SF-20520 Turku
and Pathology, 52, Finland
Turku
University,
Chickens were surgically bursectomized at 60 hr of incubation, before the bursal anlage appears. Completeness of the bursectomy was confirmed at autopsy at 10 weeks of age. These embryonically bursectomized (Bx)~ chickens are known to produce immunoglobulins of IgM, IgG, and IgA classes but so far no specific antibodies have been observed even after heavy immunization. The Bx chickens had mature plasma cells in an almost normal frequency when studied at 10 weeks of age. The amount of germinal center formation in the spleen and cecal tonsils was markedly decreased when compared to the control (Co) chickens. Also, the frequency of cytoplasmic h&positive (c-IgA+) cells was severely decreased in the Bx animals, whereas the occurrence of c-IgG+ and c-IgM+ cells was not affected to the same extent. These findings support the hypothesis that heavy-chain class switch may occur without the bursal influence, and that the bursa of Fabricius is essential only for expansion or creation of the antibody repertoire.
INTRODUCTION
Different opinions have been presented about the significance of the avian bursa of Fabricius in B-cell differentiation (1). Grossi et al. (2) detected B lymphocytes expressing simultaneously surface and cytoplasmic IgM in the bursa of Fabricius from IZday embryos lacking detectable Ig-positive cells elsewhere. Both the size of the B-lymphocyte pool and the generation of the precursors secreting different immunoglobulin classes are controlled by the bursa (3). Also, the normal switch from IgM to IgG production is thought to take place in the bursa (4). On the other hand, previous studies by us (5-7) and others (4, 8-14) provide strong evidence for the existence of extrabursal sites for B-cell development. We have found that chickens surgically bursectomized at 60 hr of incubation produce serum and surface immunoglobulins of IgM, IgG, and IgA classes, but are not able to form specific antibodies. These chickens closely resemble human patients suffering from antibody deficiency with a normal level of serum immunoglobulins (15, 16). Fitzsimmons et al. (10) reported cortical atrophy of the thymus after early I This work was supported by the Research and Science Foundation of L%ke Oy, The Emil Aaltonen Foundation, and The Finnish Cancer Union. 2 To whom correspondence should be addressed: Department of Medical Microbiology, ‘Ibrku University, SF-20520 ‘Ihrku 52, Finland. 3 Abbreviations used: Bx, bursectomized; Co, control; c-Ig’ , cytoplasmic immunoglobulin positive; LPS, lipopolysaccharide; DNPrBSA, dinitrophenyl-bovine serum albumin; PBS, phosphate-buffered saline; s-Ig+, surface immunoglobulin positive. 41 0090-1229/84 $1.50 Copyright B 1984 by Academic Press, Inc. All rights of reproduction in any form reserved
42
JALKANEN
ET Al
embryonic bursectomy and concluded that normal embryonic development of the thymus requires the presence of the bursa, and perhaps a subpopulation of T cells does not develop at all without the bursal influence. In contrast, JankoviC et al. (12) could not find any changes in the cellular makeup of the thymus following early embryonic bursectomy. Likewise, cell transfer studies have provided no evidence for the existence of bursaderived T cells (17, 18). In this work we report the existence and distribution of cytoplasmic immunoglobulin positive (c-Ig+) cells and histological findings in the spleen, cecal tonsils, Harderian gland, bone marrow, and thymus of the embryonically bursectomized (Bx) and control (Co) chickens at the age of 10 weeks. MATERIALS
AND METHODS
Chickens. Fertilized eggs from the lines P (genotype B2B2) and V (genotype B15Bt5), kept at the Department of Medical Microbiology, Tbrku University, were used. The incubation of the eggs and the care of the birds were as described previously (19). Bursectomy. Bursectomy at 60 hr of incubation was carried out according to the previously described techniques (9, 11) with some modifications. The operation was performed under aseptic conditions using microsurgery scissors just before the amniotic membrane began to cover the caudal part. The tail was cut off caudally to the leg buds. Hatchability of the operated embryos was IO-20% with 2% surviving beyond the first week of life. The controls consist of age and sex-matched normal chickens. Antigens and immunizations. The chickens were immunized at the age of 6 weeks with Escherichia coli lipopolysaccharide (LPS), levan, tetanus toxoid, and dinitrophenyl-bovine serum albumin (DNPrBSA). The origin and amount of the antigens given at a time were as follows: E. coli LPS (lipopolysaccharide W, E. coli 055:B5, Difco Laboratories, Detroit, Mich.) 10 Fg, native levan prepared from Corynebacterium levaniformis as described previously (20,21) 100 kg, commercially licensed tetanus vaccine (The National Public Health Institute, Helsinki, 10 Lf/ml tetanus toxoid) 1 ml of 1:20 dilution, and DNP,BSA, prepared according to the method of Eisen (22), 1 mg. Each antigen was administered intraperitoneally four times with l-week intervals. Autopsies and histology. For the autopsies at 10 weeks of age the chickens were sacrificed by exsanguination. To determine completeness of the bursectomy, the whole cloaca1 region was carefully dissected out, fixed with 10% neutral buffered formalin, and embedded in paraffin. Serial sections cut throughout the region were stained with van Gieson and hematoxylin and eosin. The spleen, bone marrow, one to two thymic lobes, Harderian gland, and cecal tonsils were processed in the same way and stained with van Gieson, hematoxylin-eosin, periodic acid-Schiff, and methyl- green-pyronine stains. Alcian blue mucus staining was used for samples from the Harderian gland and the cecal tonsils. Antisera. Rabbit anti-chicken y (Miles-Yeda, Rehovot, Israel) was further purified by absorption with Sepharose 4B-conjugated chicken IgM and IgA. Goat anti-chicken (Y (Miles-Yeda) and goat anti-chicken p, (Miles-Yeda) used had only
BURSECTOMY
AT 60 HOURS
OF
INCUBATION
43
anti-heavy chain specificity. Specificity of all antisera used in this study was ascertained as described previously (5). Immunoperoxidase procedure. A four-step immunoperoxidase technique was used to stain 4 p,m thick paraffin-embedded tissue sections (23). The sections were deparaffinized with xylene and alcohol and treated with pepsin for 2 hr at 37°C to release bound antigens. Endogenous peroxidase was blocked by 30 min of incubation in methanol containing 1.6% H202, and the sections were flooded with normal swine serum (Dak~immunoglobulins A/S, Copenhagen, Denmark) 1:50 for 10 min to diminish nonspecific background staining. At the first step the sections were incubated with 1:lOOO dilution of rabbit anti-chicken y, goat antichicken lo or 01(Miles-Yeda), or normal rabbit or goat serum as controls at 37°C for 1 hr. After three washings with phosphate-buffered saline (PBS) the sections were incubated with either swine anti-rabbit or anti-goat immunoglobulins (Dako) I:50 at 37°C for 30 min and then with rabbit or goat antiperoxidase-peroxidase complex (Dako) 1:500 at 37°C for 15 min. After three washings with PBS and flooding with fresh diaminobenzidinehydrogen peroxide the sections were again washed with PBS, counterstained with hematoxylin, and mounted for examination by a light microscope. In the quantitative evaluation of cells stained with anti-p, anti-y, or anti-a, the sections were first screened using a x 40 magnification; 10 fields with the highest frequency of c-Ig+ cells were chosen for further enumeration with x400 magnification. The analyses were carried out without knowing to which experimental group the subject belonged. RESULTS
All Bx birds appeared typically tailless. One out of seven Bx chickens died during immunization. The others survived in good condition to the end of the experiment. General Histology Autopsy and microscopical examination revealed no bursal remnants in the Bx chickens. Active germinal centers were observed in the spleen and cecal tonsils both in the Bx and Co groups (Fig. 1). In the cecal tonsils of the Bx chickens the number of germinal centers was only one-tenth of the Co values and in the spleen even less. Nevertheless, the number of mature plasma cells reached roughly the same level in the Bx and Co chickens except for one Bx bird (No. 2290), which had very few mature plasma cells in all the organs studied. The structure of the thymus and Harderian gland of the Bx chickens appeared normal, without any signs of atrophy. No essential difference was observed in the bone marrow sections of the Bx and Co animals. Antibody Response None of the Bx birds showed any evidence for production of specific antibodies against any of the antigens used for immunization, not even after the fourth antigenic injection. The determinations were made using enzyme-linked immu-
44
JALKANEN
ET AL
FIG. I. A germinal center (gc) surrounded by the capsule in the cecal tonsils of a Bx bird. On the left is a gland duct (gd). c-IgG+ cells stained by immunoperoxidase appear both inside and outside the germinal center. ( x 400).
nosorbent assay (5). Detailed results of these as well as production of various natural and autoimmune antibodies will be reported in a subsequent paper of this series. c-Ig + Cells c-&M. All Bx birds had c-IgM+ cells. Regarding their distribution and frequency, two findings deserve attention. (i) One of the six Bx birds (No. 2290) had very few c-IgM+ cells, whereas all the others had them in an approximately normal frequency. (ii) On average, decreased frequency of c-IgM+ cells was observed only in the bone marrow of the Bx birds: 1.9 k 0.7 cells/microscopic field compared to 3.9 k 1.1 in the Co animals (P < 0.005, Table 1). c-ZgG. The findings are very much the same as those described regarding the occurrence of c-IgM + cells. The Bx bird No. 2290 had very few c-IgG + cells, whereas on average the frequency of c-IgG+ cells was decreased only in the Harderian gland (P < 0.025) and in the bone marrow (P < 0.02, Table 2). c-ZgA. c-IgA+ cells are affected by the early bursectomy considerably more clearly than c-IgM + or c-IgG + cells. On average, c-IgA+ cells occurred in a significantly decreased frequency in the Harderian gland, spleen, and particularly in the cecal tonsils of the Bx birds, whereas for the bone marrow and thymus the
BURSECTOMY
45
AT 60 HOURS OF INCUBATION
TABLE 1 c-IgM+ CELLS IN LYMPHOID OUGANS OF INDIVIDUAL Bx CHICKENS AND OF THEIR CONTROLS(Co) AT THE AGE OF 10 WEEKS Bx chickens
Co chickens
Organ
2282
2290
2291
2292
2288
2323
Mean f SD6 Mean + SDb
Harderian gland Cecal tonsil Spleen Bone marrow Thymus
15.5~ 5.1 3.8 2.2 3.2
0.0 0.3 0.0 1.2 0.0
18.0 2.1 1.2 1.9 0.0
10.5 2.0 0.4 1.6 0.6
28.0 6.6 0.6 3.2 0.1
43.0 34.3 0.3 1.4 0.4
29.2 8.6 1.0 1.9 0.7
+ f + + *
27.1 12.8 1.4 0.7c 1.2
53.2 3.6 2.0 3.9 0.5
-t ” + t IT
19.4 1.0 3.5 1.1 0.8
a c-IgM+ cells/field ( x 400; mean of IO fields). “n=6. ’ P < 0.005, when compared to the Co group by Student’s t test,
mean Bx values were not significantly different from the corresponding Co values (Table 3). Generally, the distribution of c-Ig+ cells, both of morphologically mature plasma cells and of lymphoblasts, was similar in different tissues of Bx and Co chickens. Intracellular localization of immunoglobulins was homogeneous or fine granular with no differences between the Bx and Co animals. Both morphologically mature plasma cells and c-Ig+ lymphoblasts were found in all the peripheral lymphoid organs studied and sporadically also in the thymus. In the bone marrow all c-Ig+ cells were lymphoblasts, immature lymphocytes, and lymphocytes of medium and large size (24), both in Bx and Co chickens; no mature plasma cells were found in the bone marrow. In the Harderian gland c-Ig + cells appeared in clusters and were usually stained more intensely than in the other organs. In the spleen c-Ig+ cells were mostly seen in the mantle zones of germinal centers or surrounding them. In the cecal tonsils the c-Ig+ cells were localized in the lamina propria and germinal centers. c-Ig+ plasma cells were also observed in other parts of the intestinal mucosa. TABLE 2 c-IgG+ CELLS IN LYMPHOID ORGANS OF INDIVIDUAL Bx CHICKENS AND OF THEIR CONTROLS(CO) AT THE AGE OF 10 WEEKS Bx chickens Organ Harderian gland Cecal tonsil Spleen Bone marrow Thymus
Co chickens
2282
2290
2291
2292
2288
2323
Mean + SDb
Mean t SDb
50.7” 35.3 0.3 2.7 0.0
0.0 1.8 0.0 2.8 0.0
33.5 4.7 0.0 2.1 0.0
74.5 4.9 0.2 2.9 0.2
97.0 1.0 0.1 1.0 0.0
77.0 24.4 11.3 1.3 0.1
55.5 12.0 1.9 2.1 0.05
95.6 21.1 2.2 4.2 0.5
” c-IgG + cells/field ( x 400; mean of 10 fields). “n=6. c P < 0.025, when compared to the Co group by Student’s t test.
dP < 0.02.
2 ? ” -e 2
35.0c 14.3 4.6 o.a* 0.08
I 2 “_ + ‘-
12.9 6.7 1.8 1.6 0.8
46
JALKANEN
ET AL
TABLE 3 c-IgA+ CELLS IN LYMPHOID
ORGANS
OF INDIVIDUAL Bx CHICKENS THE AGE OF 10 WEEKS
AND OF THEIR CONTROLS
Bx chickens Organ Harderian gland Cecal tonsil Spleen Bone marrow Thymus
2282
2290
2291
2292
2288
2323
8.1” 1.4 0.1 0.4 0.0
0.0 0.0 0.0 0.5 0.1
2.8 0.2 0.0 1.6
7.1 1.0 0.0
5.0 0.9 0.0
1.1 0.0
2.0 0.0
2.9 0.5 0.0 0.3
0.1
0.0
Mean * SD6 4.3 0.7 0.02 I.0 0.03
_t t * 2 +
3.Od 0.5’ 0.04’ 0.7 0.05
(Co)
AT
Co chickens Mean 2 SDb 14.9 12.1 0.8 1.8 0.2
k -r k 5 2
5.1 5.9 0.7 0.8 0.2
a c-IgA+ cells/field (x400; mean of IO fields). bn = 6. c P < 0.025, when compared to the Co group by Student’s t test.
dP < 0.005. PP < 0.001.
In the cecal tonsils and in the Harderian gland apical parts of the epithelium were IgA positive (Fig. 2), and IgM and IgG staining of these parts was negative (Fig. 3). Also the mucus in the lumen was IgA positive, indicating active secretion of IgA.
FIG. 2. c-IgA+ cells in the Harderian gland of the Bx bird stained by immunoperoxidase. A typical plasma cell with IgA-positive cytoplasm is marked with an arrow. The apical part (two arrows) of the epithelium is stained IgA positive also. (x 1600).
BURSECTOMY
AT 60 HOURS OF INCUBATION
47
FIG. 3. c-IgM+ cells in the cecal tonsils of a Bx bird stained by immunoperoxidase. A typical plasma cell with IgM-positive cytoplasm is indicated with an arrow. Apical secretion of IgM is missing. (X looo).
DISCUSSION
The majority of c-Ig+ cells of the Bx birds had a typical plasma cell morphology, but immature lymphocytes and lymphocytes of medium and large sizes also contained c-Ig. In this respect the early bursectomy seems to have had no significant effect. The most striking effect of the bursectomy was related to the occurrence of c-IgA+ cells in the Harderian gland, cecal tonsils, and spleen. In all these organs the Bx birds had significantly fewer c-IgA+ cells than the controls (Table 3). Most probably this applies also for the thymus, where the frequency of cIgA+ cells as such is very low (Table 3). It is difficult to say whether the decrease of c-IgA+ cells in the Harderian gland, cecal tonsils, and spleen is or is not compensated by an increase of c-IgM+ cells, which occur in quite a normal frequency in these organs, but are decreased in the bone marrow (Table 1). Our previous studies have demonstrated decreased frequency of surface IgA positive (s-IgA+) cells (spleen) and of s-IgG+ cells (spleen, thymus, peripheral blood) in the Bx chickens, together with a minimal amount of serum IgG (5, 6). The decreased occurrence of germinal centers in the Bx birds is a remarkable finding. However, it is related to the defect in antibody formation and IgG production found typically in the Bx birds (5-7). Otherwise, no structural changes, hypoplasia, or atrophy were observed in the peripheral lymphoid organs of the
48
JALKANEN
ET AL
Bx birds at 10 weeks of age. Fitzsimmons and co-workers (IO, 25) found marked cortical hypoplasia of the thymus in the chickens bursectomized as early embryos. It is possible that lymphoid structure of the thymus may develop more slowly in Bx than in normal chickens. This can partially explain the discrepancy with our observations, since Fitzsimmons et al. mostly analyzed their birds as embryos. Our findings are in concordance with those of Jankovic et al. (12), who found no changes in the cellular makeup of the thymus in the chickens bursectomized at 52-64 hr of incubation. On this basis, the bursa of Fabricius seems not to have any marked effect on the structure of other lymphoid organs. At the present, a meaningful synthesis of the findings concerning Bx chickens seems difficult. A summary of the occurrence of c-Ig + and s-Ig + cells is presented in Table 4. Data on the thymus have been excluded, due to the low frequency of B cells or their precursors in this organ and to the uncertainties caused by it in the interpretation. Likewise, in the spleen the statistical significance of the decreased frequency of cells with c-IgM, s-IgM, or c-IgG is hampered either by large individual variations or by low values. It is clear, however, that c-Ig’ and s-Ig+ cells of all three isotypes are found in the Bx birds, indicating, as also stated before (5-7), that switch in isotype expression occurs independently of the bursa of Fabricius. It is possible that Bx birds have a disturbance in the regulation of the switch since the occurrence of cellular IgA and IgG seems to be affected by the early bursectomy more than that of cellular IgM, particularly in the peripheral lymphoid tissues. In addition, Bx chickens are known to be unable to produce specific antibodies in spite of immunoglobulin production (5, 6). These findings have led us to suggest that the bursa of Fabricius is not essential for the isotype switch, but rather only for the expansion or creation of the antibody repertoire. The explanation for the occurrence of immunoglobulins without antibody specificity remains to be found. One could argue that the bursectomies were subtotal and a few follicles which were sufficient for isotype switch, but not for the genTABLE FREQUENCY
OF c-Ig+
CELLS BASED
AND
s-Ig+
ON DATA
CELLS
4
IN DIFFERENT
PRESENTED
IN THIS
AND
ORGANS
OF lo-WEEK-OLD
A PREVIOUS
Bx
CHICKENS.
PAPERY
Cells Organ Harderian gland Cecal tonsils Spleen Peripheral blood Bone marrow
c-IgM +
s-&M +
c-&G+
s-IgG +
Normal Normal Normal ND JJ
NDb ND Normal 11 t
i’ Normal Normal ND -1
ND
c-IgA + l’J”le
rNl? JJJ Normal
:i~ Normal
s-IgA + ND ND i-1 Normal Normal
n S. Jalkanen, K. Granfors, M. Jalkanen, and P. Toivanen, J. Immunol. 130, 2038, 1983. b Not determined. c J Slightly decreased ( t slightly increased); P 5 0.05 when compared to normal controls by Student’s t test. d & i Decreased; P 5 0.01. e 4 4 J Markedly decreased; P 5 0.001.
BURSECTOMY
AT 60 HOURS OF INCUBATION
49
eration of antibody diversity, developed. However, this possibility was excluded by careful microscopic examination. All of the chickens studied appeared completely tailless, lacking also parts of the coprodeum, the most cranial part of the avian cloaca. These findings also exclude the possibility that a few bursal follicles developed after the bursectomy and were already in a complete involution at the time of the autopsy. There were birds in which microscopic examination revealed a few bursal follicles. However, all of these birds had a considerable part of the cloaca and a clearly demonstrable antibody response against all the antigens used; these chickens were excluded from the study. In the bone marrow, decrease of c-IgM+ cells and increase of s-&M+ probably argues for a disturbance in the maturation of B cells. In spite of this, bone marrow appears more normal than the other tissues studied regarding the occurrence of c-Ig+ or s-Ig+ cells. Bone marrow cells of Bx birds respond in a normal fashion by cell proliferation to anti-p or anti-y, whereas spleen cells of Bx birds respond significantly less than spleen cells of normal chickens (7). However, it is obvious that more definite explanations for these peculiarities have to wait for molecular analyses in experimental animals or human patients (16). ACKNOWLEDGMENTS We thank Dr. Lauri Pelliniemi for advice in immunoperoxidase methodology, Soile Niittoaho, Tarja Puurunen, Marjo Vesanto, and Anja Vuoristo for expert technical assistance, and Eija Nordlund for secretarial help.
REFERENCES I. Toivanen, A., Toivanen, P., Eskola, J., and Lassila, O., In “Avian Immunology” (M. E. Rose, L. N. Payne, and B. M. Freeman, Eds.), pp. 45-62, British Poultry Science Ltd., Edinburgh, 1981. 2. Grossi, C. E., Lydyard, P. M., and Cooper, M. D., J. Immunol. 119, 749, 1977. 3. Lawton, A. R., Kincade, P. W., and Cooper, M. D., Fed. Proc. 34, 33, 1975. 4. Subba Rao, D. S. V., McDufie, F. C., and Glick, B., J. Immunol. 120, 783, 1978. 5. Granfors, K., Martin, C., Lassila, O., Suvitaival, R., Toivanen, A., and Toivanen, P., Clin. Immunol.
6. 7. 8. 9. 10. 11. 12. 13. 14. IS. 16. 17. 18.
Immunopathol.
23, 459, 1982.
Jalkanen, S., Granfors, K., Jalkanen, M., and Toivanen, I?, J. Immunol., 130, 2038, 1983. Eerola, E., Jalkanen, S., Granfors, K., and Toivanen, A., J. Zmmunol., 131, 120, 1983. Lemer, K. G., Glick, B., and McDufEe, F. C., J. Immunol. 107, 493, 1971. Fitzsimmons, R. C., Garrod, E. M. F., and Garnett, I., Cell. Immunol. 9, 377, 1973. Fitzsimmons, R. C., Dixon, D. K., and Kocal, E. M. F., In “Developmental Immunobiology” (J. B. Solomon and J. D. Horton, Eds.), pp. 387-394, Elsevier/North-Holland, Amsterdam, 1977. Jankovic, B. D., Knezevic, Z., Isakovic, K., Mitrovic, K., Markovic, B. M., and RajCeviC, M., Eur. J. Immunol. 5, 656, 1975. Jankovic, B. D., Isakovic, K., Markovic, B. M., RajCeviC, M., and Knezevic, Z., Exp. Hematol. 4, 246, 1976. Jankovic, B. D., Isakovic, K., Markovic, B. M., and RajCeviC, M., Immunology 32, 689, 1977. Sato, K., and Abe, S., Immunology 28, 293, 1975. Rothbach, C., Nagel, J., Rabin, B., and Fireman, P., J. Pediatr. 94, 250, 1979. Saxon, A., Kobayashi, R. H., Stevens, R. H., Singer, A. D., Stiehm, E. R., and Siegel, S. C., Clin. Immunol. Immunopathol. 17, 235, 1980. Lassila, O., Acta Pathol. Microbial. &and. [C’j 87, 287, 1979. Weber, W. T., and Alexander, J. E., J. Zmmunol. 121, 653, 1978.
50
JALKANEN
ET AL
19. 20. 21. 22. 23.
Toivanen P., and Toivanen, A., Eur. J. Immunol. 3, 58S, 1973. Miranda, J. J., Immunology 23, 829. 1972. Moreno, C., Courtenay, B. M., and Howard. J. G., fmmu~oc~ernisf~ 13, 429, 1976. Eisen, H. N., Belman, S., and Carsten M. E.. J. Amer. Chem. Sot. 75, 4.583. 1953. Sternberger, L. A., Hardy, P. H.. Jr., Cuculis, J. J., and Meyer, H. G., J. Histochem. Cytochem 18, 315, 1970. 24. Lucas, A. M., and Jamroz, C., “Atlas of Avian Hematology,” Agriculture Monograph 25. tJ.S Department of Agriculture, Washington, D.C., 1961. 25. Dixon, D. K.. and Fitzsimmons, R. C., Dev. Comp. Immunol. 4, 713, 1980. Received February 8, 1983; accepted with revision May 31, 1983.