In vitro organ culture of embryonic bursa of fabricius

Developmentaland ComparativeImmunology,Vol. 14, pp. 239-246, 1990 Printed in the USA. All rights reserved.

0145-305X/90 $3.00 + .00 Copyright © 1990 Pergamon Press plc

IN VITRO ORGAN CULTURE OF EMBRYONIC BURSA OF FABRICIUS Jerzy S. Nowak*, Olli Vainiot,$, and Olli Lassilat Basel Institutefor Immunology,Basel, Switzerland (Submitted March 1988;Accepted June 1988) [3Abstract--An in vitro method for organ culture of embryonic bursa of Fabricius is presented. It is shown that bursa cells proliferate in in vitro culture as evidenced by [3H]-thymidine incorporation. We assessed the expression of B-cell alloantigen (Bu-la and Bu-lb), class I (B-F) and class II (B-L) antigens of chicken major histocompatibility complex (MHC), and surface immunoglobulin (sIg) on cultured bursa cells using specific monoclonal antibodies (mAb). Cells from 13-day and 14-day embryonic bursae incubated in organ culture for l to 2 weeks developed characteristic patterns of surface phenotype observed in adult chicken except for B-F antigen, whose expression was much lower than in vivo. These resuits indicate that the maturation of bursa cells in organ culture follows the in vivo development, except for the expression of MHC class I antigens. Furthermore, we demonstrate the in vitro repopulation of bursae from cyclophosphamide(cy)-treated chickens by cells from Bu-I antigen disparate normal bursae. RKeywords--Organ culture; Bursa of Fabricius; B-cell development; Chicken. Introduction

The bursa-dependent B-cell system in the chicken offers a good model to study ontogenetic development of B-lympho* On leave from the Institute of Human Genetics, Polish Academy of Sciences, Poznan, Poland. f Present address: Department of Medical Microbiology, Turku University, Turku, Finland. $ Address correspondence to Dr. Olli Vainio, Department of Medical Microbiology, Turku University, SF-20520 Turku, Finland.

cytes. Studies on bursa cell development have demonstrated that differentiation of chicken B-lymphocytes is accompanied by the acquisition of specific cell surface markers, like slg, Bu-I and B-L antigens and function (I-5). Recently, production of specific mAb has made it possible to investigate more precisely the ontogenetic d e v e l o p m e n t of B cells in the chicken (1,6-9). The organ culture method for mouse thymus, described initially by Auerbach (10), has been used subsequently by many workers to study various aspects o f t h y m u s d i f f e r e n t i a t i o n in v i t r o (11-13). Mouse T-cell differentiation in vitro seems to recapitulate the in vivo differentiation. Organ culture of embryonic bursa would offer a physiological approach to study early stages of B-cell differentiation in vitro in the intact bursal microenvironment. This would enable us to dissect more precisely various requirements for B-cell differentiation and also allow studies like in vitro virus transformation of B-cell precursors by various avian retroviruses. The present study was undertaken to establish the conditions for bursal organ culture. Using a panel of mAb directed to surface antigens of chicken B-cells, we have demonstrated that bursal maturation in organ culture seems to parallel the developmental sequences observed in vivo. Furthermore, we have demonstrated the in vitro colonization of Bcell-depleted bursae from cy-treated chickens by normal bursa cells. These findings should be useful in further studies of normal and malignant B-cell

239

240

development in the chicken bursa of Fabricius.

Materials and Methods

J.S. Nowak et al.

tibody (19) provided by Dr. E. Houssaint, University of Nantes, France. M-4 was used as 1:100 diluted ascitic fluid and all other antibodies were used as undiluted hybridoma culture supernatant.

Animals Chickens used in these studies were maintained at Roche, Gipf-Oberfrick, Switzerland. The partly inbred H.B21 strain was derived from a Hyline FP ([B Is × B 21] F 1 hybrid) stock by more than 10 generations of selection for chickens carrying the B21 major histocompatibility complex (MHC) haplotype (14). The H.B21 strain has been divided into two sublines, distinguishable by the presence of Bu-la or Bu-lb alloantigen on their B-lymphocytes. The H.B21a subline expresses the Bu-la antigen which is detected by mAb L22 (8) and 21-1A4 (9). The H.B21b subline expresses the Bu-lb antigen detected by mAb 5-11G2 (9). The H.B21b subline corresponds to the former subline H.B21 called Bu-la-negative (15). The highly inbred CB line originated from the Institute of Molecular Genetics, Czechoslovak A c a d e m y of Sciences, Prague (16). All fertilized eggs were incubated at 38°C and embryonic stages were determined by the duration of the incubation.

Monoclonal Antibodies 21-1A4 and 5-11G2 recognizing Bu-la and Bu-lb B cell alloantigen, respectively, have been described elsewhere (9). M-4 antibody (1) was used to detect cells expressing surface IgM and was purchased from Southern Biotechnology Assoc. Inc., Birmingham, AL. Expression of chicken MHC class I (B-F) and class II (B-L) molecules was determined by using 37C.18 (17) and 21-1A6 antibodies (18), respectively. Epithelial cells of the bursa were detected by BEP-I an-

Cyclophosphamide Treatment A freshly prepared solution of cy (Farmos Group Ltd., Finland), at a dose of 4 mg per chick, was injected intraperit o n e a l l y on four c o n s e c u t i v e days starting on the day of hatching (20).

Organ Culture Bursae removed from 10-17 day inc u b a t e d e m b r y o s , and from n e w l y h a t c h e d normal and 1-week-old cytreated birds were cultured essentially according to the procedure described in detail for mouse thymus organ culture (11-13). Bursae of 10-14-day incubated embryos were cultured as whole organs, while 17-day and older embryo bursae were cut into small pieces. Four to six individual bursae or bursa pieces were placed on the surface of 0.8 Ixg (¢ 13 mm) polycarbonate filters (Nucleopore Corp., Pleasanton, CA) supported by strips of gelatin sponge (Sterispan, Allen and Hanburys Ltd., London, England) soaked in culture medium. For bursa organ cultures, as well as for chicken cell cultures, we used modified Iscove's modified Dulbecco's medium (IMDM), described in detail elsewhere (21), supplemented with 0.5-1% normal chicken serum and 10% fetal calf serum. The organ cultures were carried out in 9-cm Petri dishes filled with 5 ml medium or in 6-well tissue culture clusters (Costar, Cambridge, MA) with 2 ml medium in each well. The medium was changed every 2 - 4 days. The cultures were grown in a fully humidified incubator in 5% CO2 in air at 37°C.

241

Organ culture of the bursa

Colonization of Cy-Treated In Vitro

Bursa

One piece (l/4-1/2 of the organ) of bursa from cy-treated 7-day-old birds (cy-bursa) was cultured in contact with normal bursa (l/6- l/8 of the organ) of a newly hatched chicken. Normal and cybursae were of the opposite Bu-1 allotype (from the strains H.B2la vs. H.B2lb) enabling us to distinguish between the lymphocytes from the two bursae. All culture conditions were the same as described above. The cultures were terminated after 9 days and the cells from bursae were analyzed for surface phenotype. The organs were also assessed for [3H]-thymidine incorporation.

Cell Proliferation Assay Cell proliferation was measured by [‘HI-thymidine incorporation. Organ cultures were pulsed with 5 @i/ml of [3H]-thymidine for 5 h. Then the organs were placed on filter strips (Whatman GF/A) between two strips of fine nylon mesh and squeezed thoroughly with the tips of bent forceps. Afterwards the obtained cells were washed and harvested manually on filter paper strips using a filter glass unit and counted in a liquid scintillation counter. Approximately 200 ml of distilled water was used to wash the harvested cells. As a control, the organs were transferred to cold medium, incubated on ice for 5 h and then manually harvested as described above. The control organs usually showed radioactivity not exceeding 500 c.p.m.

Preparation

of Cell Suspensions

Embryonic and in vitro cultured bursae were cut with scissors into small

pieces, placed in Earl’s balanced salt solution (EBSS) between two strips of fine nylon mesh and squeezed by scraping the upper mesh with the tip of a syringe plunger. After washing, cells were resuspended in phosphate-buffered saline containing 5% fetal calf serum and 0.1% sodium azide (PBS-FCS-NaN,). Cell viability (determined by trypan blue exclusion test) for freshly isolated embryonic bursa cells was 50-70% and for organ culture bursa cells in a range of 40-60%.

Zmmunojluorescence

Staining

Cells were stained by using an indirect immunofluorescence technique. All incubations and washings were carried out at 4°C in PBS-FCS-NaN,. Twenty ~1 of bursa cells at a concentration of 15 x lo6 cells/ml were incubated with 30 ~1 of appropriate mAb for 30 min in a 96-well tissue culture cluster. The cells were washed three times and incubated for an additional 30 min with 1: 100 diluted FITC-labeled-F(ab’), fragment of sheep anti-mouse Ig (Silenus Laboratories, Hawthorn, Australia). Then the cells were washed four times and examined under incident illumination on a LeitzWetzlar Orthoplan fluorescence microscope. Usually 200-400 cells were scored.

Results Growth of the Bursa in Organ Culture

Bursae removed from 17-day incubated chicken embryos were cut into small pieces (approx. l/4 of the size) and cultured in vitro for various time periods up to 14 days. In the consecutive culture days [3H]-thymidine incorporation was estimated. The [3H]-thymidine uptake was always the highest after l-day culture, then decreased and was maintained approximately at the same level up to 14

242

J. S. Nowak et

days when the cultures were terminated (Table 1). By the end of the first day in organ culture many bursa cells died which can also be observed in the steep decline of the [3H]-thymidine incorporation. Thereafter a stable DNA synthesis and cell proliferation was seen during the next two weeks with a slight decline towards the end of the culture period. In subsequent experiments the proliferation of epithelial cells in bursal organ cultures was evaluated and found to be minimal, since the percentage of cells stained with a mAb specific in bursal epithelial cells (BEP-1) remained very low in cultures lasting up to 14 days. Table 2 presents the [3H]-thymidine incorporation by bursae from embryos of different ages cultured in vitro for 7 days. It should be noted that in bursa organ cultures a great variation of the 3H-thymidine uptake was observed. High range is observed especially in 13day-old embryonic bursae cultured for 7 days where [3H]-thymidine uptake ranged from 6.996 c.p.m. up to 34.225 c.p.m.

Surface Phenovpe Bursa Cells

of Organ Cultured

Thirteen embryonic bursae were tured in vitro for 7 days whereafter expression of surface antigens was lyzed. Using a panel of mAb it shown that the frequency of bursa

culthe anawas cells

Table 1. PHI-Thymidine incorporation by 17-day embryonic bursa cultured for various periods in in vlfro organ culture Culture Period (days) 1 2 3 4 7 14

[3H]-Thymidine Incorporation Mean c.p.m. 2 SD 39210 13330 13305 13871 9075 8940

2 18281 2 2045 f 2208 k 2759 AZ3446 ? 3777

No. of Experiments 8 8 6 6 9 6

al.

Table 2. PHI-Thymidine incorporation by embryonic bursa cultured in vitro for 7 days

Age of Embryo 10 12 13 14 17

day day day day day

[3H]-Thymidine Incorporation (Mean c.p.m. -+ SD) 5181 8759 15900 13092 9075

k r k 2 f

2632 5417 12464 1367 3446

No. of Experiments 2 14 4 4 9

expressing Bu-lb alloantigen increased significantly during the 7-day culture reaching almost the level seen in the control 20-day embryonic bursa (Table 3). The expression of sIg by embryonic bursa cells was evaluated with a polyvalent antibody to chicken Ig. As shown in Table 3, 13-day embryonic bursa cultured in vitro for 7 days expressed sIg in much higher frequencies than detected in 13-day embryonic bursa and little less than in 20-day embryonic bursa. B-L, the class II antigen of chicken MHC, which is detected by mAb 21-lA6 was expressed on 19% of 13-day embryonic bursa cells and increased during 1 week of culture to 29% The percentage of 20-day bursa cells expressing B-L antigen was higher than that observed in organ-cultured bursa. B-F, the class I antigen of chicken MHC detected by 37C.18 antibody, is fully expressed on bursa cells after hatching but stained weakly the embryonic bursa cells. Because of the weak fluorescence only a rough estimation of B-F-positive cells can be given (Table 3). The expression of B-F antigen was much higher on cultured bursa cells than that detected on 13-day embryonic bursa, but lower than on 20-day embryonic bursa cells. This holds true for other antigens tested and suggests that, in in vitro culture, the appearance of specific surface markers takes place slower than in vivo or that antigen density per cell is lower. BEP-1 antibody was always used to test the frequency of epithelial cells. It is likely that epithelial cells along with

Organ culture of the bursa

243

Table 3. Expression of surface antigens on 13-day embryonic bursa cells cultured for 7 days in organ culture Normal Embryonic Bursa Surface Antigen

13-Day Embryonic Bursa Cultured for 7 Days In Vitro

13 days

20 days

Bu-lb Ig B-L B-F* BEP-It

85 66 29 30 0

6 2 19 2 2

90 81 41 50 0

Figures represent percentage of stained cells. * Because of weak fluorescence only a rough estimation is given. t Lymphocytes were totally negative, but a few nonlymphoid cells were positively stained.

other bursa cells proliferate in in vitro organ culture. Therefore, it was necessary to check the contribution of epithelial cells in these cultures. After a 3week culture only 4% of the cells stained with BEP-I a n t i b o d y . In a d d i t i o n , usually a few cells per slide, morphologically nonlymphoid-like, stained with BEP-I antibody. These cells were excluded from the counting of the cells expressing other markers.

In Vitro Colonization of Cy-Bursa To determine whether the bursa cells in organ culture are able to colonize cybursa, a coculture of normal and cybursa was set up. Donors of the normal and cy-bursa differed at the Bu-I allotype. The bursae were allowed to grow in culture for 9 days and afterwards the cells were stained with anti-Bu-la and anti-Bu-lb mAb (Table 4). It appeared Table 4. In vitro colonization of cy-bursa by normal bursal cells during a 9-day organ culture Bu-I AIIotype of

Bu-I Antigen Expression

Normal Bursa

Bu-lb

Bu-la

b

a

0:0

46:54

b

--

0

0

--

a

0:0

86:90

Cy-Bursa

Results are expressed as percentage of positive cells from two experiments.

that cy-bursae were repopulated by cells from normal bursae of the opposite Bu-I allotype. Approximately half of the cells in recolonized cy-bursae express Bu-la antigen. Cells with slg and B-L (class II) antigen were also detected (data not shown). Cells expressing Bu-lb antigen were totally absent in the repopulated bursae. BEP-I antibody stained only a few nonlymphoid cells. In control cybursa only cells expressing B-L antigen were present, which most likely are of macrophage origin. Normal bursa cells from newly hatched chickens cultured in vitro for 9 days expressed around 90% of Bu-la positive cells.

Discussion The presented results indicate that chicken embryonic bursa cells can grow and differentiate in an organ culture in vitro. Furthermore, we demonstrate in vitro colonization of cy-bursa by normal bursa cells. The culture conditions for growing embryonic bursae were very similar to those described for organ culture of fetal mouse thymus (13). In a series of preliminary experiments it was found that for optimal growth of the bursa in culture, fetal calf serum is necessary. In addition the temperature must be 37°C instead of 40°C, usually used for culture of chicken T-lymphocytes.

244

Growth of the bursa in in vitro organ culture was monitored by [3H]-thymidine uptake. Many authors have counted the cells for estimation of cell proliferation in mouse thymus organ culture and compared that with the original number of cells per one thymus lobe (11-13). Since small pieces of bursa for the organ cultures were used it was difficult to estimate precisely the exact increase in cell number. Therefore, in this study, the [3H]-thymidine uptake was used for determination of cell proliferation. From the results obtained (Tables 1 and 2) it is evident that the cells do proliferate in the bursa organ culture as measured by DNA synthesis. The growth of mouse thymus in organ culture is due to the proliferation of both epithelial and lymphoid cells (12). It is most probable that these two cell populations are also involved in the bursal growth in vitro. The contribution of epithelial cells to the bursal growth in organ culture was tested by staining single cell suspensions with BEP-I mAb which specifically recognizes epithelial cells (19). The content of the epithelial cells in 1-2 week cultures was always very small. Up to 4% of round nonlymphoid-like cells and a few cells of irregular shape were positively stained with BEP-I antibody. It can be concluded that the [3H]-thymidine was incorporated mainly by proliferating lymphoid cell. Bursae from 13-day and 14-day-incubated embryos were used to study the expression of cell surface markers in organ culture. At that embryonic stage bursa colonization by B-cell precursors is almost completed (2,3). A low frequency of cells expressing slg, Bu-1, B-L and B-F antigens was detected in 13day-embryonic bursa (Table 3). These results are consistent with the described appearance of these cells in the developing bursa (1-4). It has been suggested that the bursa itself does not serve as an inducive environment for expression of B-L and Bu-I antigens but rather enables

J.S. Nowak et al.

the development of cells already preprogrammed to become mature B-lymphocytes (24). Taking this into account, we have studied the development of surface phenotype of bursa cells in organ culture, starting from the 13th to 14th day of incubation, when bursa cells start to proliferate and differentiate in the follicles. In 7-day culture the frequency of cells expressing Bu-lb antigen increased up to 85% of the level of 20-day embryonic bursa (Table 3). sIg-positive cells were found after 1 week of culture at somewhat lower frequency than in vivo, whereas in 3-week culture the values were similar (Table 3 and unpublished results). It is possible that after one week in culture the density of sIg on bursa cells is lower than in vivo. Following 7-day or 3-week culture the bursa cells expressing sIg and Bu-lb antigen significantly increased, which indicates that maturation process of B-lymphocytes is taking place during in vitro culture. Further evidence for B-cell maturation was achieved by analysis of B-L and B-F antigen expression. B-F, the class I antigen of chicken MHC, showed relatively low expression on embryonic bursa cells. Usually the fluorescence produced with anti B-F antibody is very weak on embryonic cells. In control 2week-old chicken the fluorescence was very strong and almost 100% of bursa cells were B-F antigen-positive (data not shown). In bursa organ culture we have never observed full expression of B-F antigen and the percentage of B-F-positive cells was much lower than in 2week-old chickens. 21-1A6 mAb, used for detection of B-L-antigen, stained 43% of bursa cells from 2-week-old chickens. Among B-L-positive cells about 50% showed very strong fluorescence. The frequency of B-L-positive bursa cells is similar to the frequencies reported earlier by the use of this mAb (18). In 1- to 3-week organ culture the percentage of cells expressing B-L an-

Organ culture of the bursa

245

tigen was similar to that seen in appropriate controls (Table 3 and unpublished results). This implies that B-L-antigen expression is generated in vitro culture similarly as in vivo. On the basis of these results it can be concluded that in vitro organ cultured bursa cells develop the characteristic pattern of surface phenotype seen in young chicken, except for B-F antigen. In vivo allogeneic reconstitution of cy-bursa is a well established method of studying B-cell differentiation, T-B-cell collaboration and transplantation tolerance (20,22,23,25). Using chickens of opposite Bu-I allotype enables easy distinction of cells from donor or recipient origin. Nine-day in vitro cultured cybursa of Bu-lb allotype was repopulated by normal Bu-la bursa crlls. This was evident from the determination of surface phenotype of cells residing in the cy-bursa after 9-day culture (Table 4). It should be noted that cy treatment causes

complete removal of slg- as well as Bu-lpositive lymphoid cells from the bursa. Therefore, the presence in cy-bursa of lymphocytes expressing slg or Bu-la antigen suggests the in vitro colonization process. In conclusion, we have described in vitro organ culture system, which allows us to study early events of B-cell differentiation in the chicken more precisely. The possibility of colonizing a cy-bursa in vitro by normal bursa cells should be of great value for further studies on Bcell development as well as on B-cell function.

Acknowledgements--We thank

B. Riwar and B. Walch for technical assistance, J. Hossmann and E. Nordlund for typing the manuscript and Drs. P. Matzinger and P. Kisielow for critical reading of the manuscript. The Basel Institute for Immunology was founded and is fully supported by E Hoffmann-La Roche & Co., Ltd., Basel, Switzerland.

References 1. Chen, C. H.; Cooper, M. D. Identification of cell surface molecules on chicken lymphocytes with monoclonal antibodies. In: Toivanen, A; Toivanen, P., eds. Avian Immunology: Basis and Practice. Vol. 1. Boca Raton, FL: CRC Press Inc.; 1987; pp. 137-154. 2. Pink, J. R. L.; Lassila, O.; Vainio, O. B-lymphocytes and their self-renewal. In: Toivanen, A.; Toivanen, P., eds. Avian Immunology: Basis and Practice. Boca Raton, FL: CRC Press Inc.; 1987; pp. 65-78. 3. Houssaint, E. Cell lineage segregation during bursa of Fabricius ontogeny. J. Immunol. 138:3626-3634; 1987. 4. Ewert, D. L.; Cooper, M. D. Ia-like alloantigens in the chicken: serologic characterization and ontogeny of cellular expression. Immunogenetics 7:521-535; 1978. 5. Weiil, J.-C.; Reynaud, C.-A. The chicken B cell compartment. Science 238:1094-1098; 1987. 6. Pink, J. R. L.; Gilmour, D. G. Surface antigens of avian blood cells. In: Miyasaka, M.; Trnka, Z., eds. Differentiation Antigens in Lymphohemopoietic Tissues. New York: M. Dekker Inc.; 1988; pp. 361-385. 7. Ewert, D. L.; Munchus, M. S.; Chen, C. H.; Cooper, M. D. Analysis of structural properties and cellular distribution of avian Ia antigen by using monoclonal antibody to monomor-

phic determinants. J. Immunol. 132:25242530; 1984. 8. Pink, J. R.; Rijnbeek, A. M. Monoclonal antibodies against chicken lymphocyte surface antigens. Hybridoma 2:287-296; 1983. 9. Veromaa, T.; Vainio, O.; Eerola, E.; Toivanen, P. Monoclonal antibodies against chicken Bu-la and Bu-lb alloantigens. Hybridoma 7:41-48, 1988. 10. Auerbach, R. Morphogenetic interactions in the development of the mouse thymus gland. Dev. Biol. 2:271-281; 1960. 11. Mandel, T. E.; Kennedy, M. M. The differentiation of murine thymocytes in vivo and in vitro. Immunology 35:317-331; 1978. 12. Robinson, J. H.; Jordan, R. L. Thymus in vitro. Immunol. Today 4:41-45; 1983. 13. Kisielow, P.; Leiserson, W.; von Boehmer, H. Differentiation of thymocytes in fetal organ culture: analysis of phenotypic changes accompanying the appearance of cytolytic and interleukin-2 producing cells. J. Immunol. 133:1117-1123; 1984. 14. Briles, W. E.; Bumstead, N.; Ewert, D. L.; Gilmour, D. G.; Gogusev, J.; Hala, K.; Koch, C.; Longenecker, B. M.; Nordskog, A. W.; Pink, J. R. L.; Schierman, L. W.; Simonsen, M.; Toivanen, A.; Toivanen, P.; Vainio O.; Wick, G. Nomenclature for chicken major his-

246

15.

16.

17.

18.

19.

tocompatibility (B) complex. Immunogenetics 15:441-447; 1982. Pink, J. R. L.; Vainio, O.; Rijnbeek, A.-M. Clones of B lymphocytes in individual follicles of the bursa of Fabricius. Eur. J. Immunol. 15:83-87; 1985. Hagek, M.; Knizetov,-l, E; Mervartov~, H. Syngenic lines of chickens. I. Inbreeding and selection by means of skin allografts and tests for erythrocyte antigens in C line in chickens. Folia Biol. (Praha) 12:335-342; 1966. Pink, J. R. L.; Kieran, M. W.; Rijnbeek, A. M.; Longenecker, B. M. A monoclonal antibody against chicken MHC class I (B-F) antigens. Immunogenetics 21:293-297; 1985. Veromaa, T.; Vainio, O.; Jalkanen, S.; Eerola, E.; Granfors, K.; Toivanen, P. Expression of B-L and Bu-1 antigens in chickens bursectomized at 60 h of incubation. Eur. J. Immunol. 18:225-230; 1988. Houssaint, E.; Diez, E.; Hallet, M.-M. The bursal microenvironment phenotypical characterization of the epithelial component of the bursa of Fabricius with the use of monoclonal antibodies. Immunology 58:43-49; 1986.

J.S. Nowak et al.

20. Toivanen, E; Toivanen, A. Bursal and postbursal stem cells in chicken. Functional characteristics. Eur. J. Immunol. 3:585-595; 1973. 21. Lassila, O.; Alanen, A.; Lefkovits, l.; Cooper, M. D.; Pink, J. R. Immunoglobulin diversification in embryonic chicken bursa and in individual bursal follicles. Eur. J. Immunol. 18:943-949; 1988. 22. Vainio, O.; Toivanen, A. Cellular cooperation in immunity. In: Toivanen, A.; Toivanen, E, eds. Avian Immunology: Basis and Practice. Vol. 2. Boca Raton, FL: CRC Press Inc; 1987; pp. 1-12. 23. Lehtonen, L.; Vainio, O.; Toivanen, E Lymphoid cell chimerism and transplantation tolerance induced by bursal and postbursal cells. Transplantation 41: 398 - 405; 1985. 24, Ratcliffe, M. J. H.; Lassila, O.; Pink, J. R. L.; Vainio, O. Avian B precursors: surface immunoglobulin expression is an early, possibly bursa-independent event. Eur. J. lmmunol. 16:129-133; 1986. 25. Lassila, O.; Vainio, O.; Matzinger, E Do B cells turn on virgin T cells? Nature 334:253255; 1988.