- Email: [email protected]
Neonatal tolerance induction in the thymus to MHC-class II-associated antigens IV. Significance of intrathymic chimerism of blood-born la+ cells in Mls tolerance
CELLULAR
IMMUNOLOGY
136.373-387
(199
1)
Neonatal Tolerance Induction in the Thymus to MHC-Class II-Associated Antigens IV. Significance of lntrathymic Chimer&m of Blood-Born la+ Cells in MIS Tolerance MASAMICHI HOSONO,* MAFUMI KUROZUMI,~ MUNEO INABA,$ SHIN IDEYAMA,*‘§ MITSURU TOMANA,* JUN-ICHIRO GYOTOKU,* YOSHIMOTO KATSURA,* AND TOMOHIDE HOSOKAWA~~ Departments of *Immunology and of tClinical Laboratory, Chest Disease Research Institute, and $Department of Dermatology, Faculty of Medicine, Kyoto University, Kyoto, Japan; $Department of Pathology, Kansai-Medical University, Moriguchi, Osaka, Japan; and I’Department of Preventive Medicine, Kyoto Prefecture University of Medicine, Kvoto, Japan Received
November
6, 1990;
accepted
April
2, 1991
The significance of thymus cell chimerism in the induction and maintenance of tolerance was investigated. MIS- I b BALB/c mice were neonatally tolerized by the intravenous administration of either bone marrow (BM) cells or peritoneal cavity (PerC) cells from MIS-lb’” (BALB/c X AKR) F, mice. Tolerance was long-lasting in the BM cell group, but transient in the PerC cell group, probably because PerC cells lack hemopoietic stem cells required for a continuous supply of tolerance-inducing cells. The degree of anti-Mls-la responsiveness of these BALB/c thymus cells was correlated with the degree of intrathymic distribution of the inoculated F, cells. The effect of BM cell inoculation, resulting in a year-long deletion of MIS-la-reactive VB6-bearing T cells is in marked contrast to that of PerC cell inoculation which causesonly a transient loss of V,6+ mature thymocytes (for about 1 week after birth). This functional profile of the tolerant state correlates well with the degree and persistence of the intrathymic presence of F, type Ia+ cells. The longlasting presence of donor-derived cells throughout the thymus tissue in the BM cell group is also in marked contrast to the early disappearance of Ia+ cells (within 2-3 weeks) from the cortex and then from the medulla in the PerC cell group, although these Ia’ cells were once spread throughout the thymus tissue 4 days after the tolerance-inducing cell inoculation. Taken together with a failure to induce consistent unresponsiveness to Mls-1” determinants in Mls-lb thymocytes regenerating in MIS-la-thymic epithelial environments, all the above data indicate that intrathymic chimerism caused by hemopoietic stem cell-derived MHC-class II-bearing cells is a requisite for the induction and maintenance of unresponsiveness by means of clonal deletion in experimentally as well as naturally induced tolerance to MIS determinants. o 1991 Academic ores. IIIC.
INTRODUCTION The thymus is a central organ for differentiation, maturation, and antigen-repertoire generation of T cells, especially of the self-major histocompatibility complex (MHC)class II-restricted helper/inducer subsets ( l-5). In the generation of the T cell repertoire of antigen recognition, thymus T cells directed to self-MHC-class II-associated antigens and possessing autoreactive potentiality are thought to be eliminated by a negative 373 0008-8749/91 $3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
374
tli)SOlrt) t- i .iL
selection process before their peripheralization. Since the pioneer studies of Bilhngham, Brent. and Medawar (6). immunological tolerance induced by the administration of semiallogeneic hemopoietic stem (HPS) cells into newborn mice has been widely employed to clarify the fail-safe mechanisms in the immune system. However. the mechanisms involved are not yet completely clear. In order to investigate the celluiar events in the negative seiechon process in the developing thymus, we employed a tolerance induction system, injecting intravenously hematolymphoid cells of Mls-semiallogeneic F, mice into neonates. The tolerance thus induced in the thymus as early as 1 week after birth was specific to the corre-, sponding antigens (7-9) and closely associated with MHC antigens (8). The effective cells for the induction of tolerance are thought to be of HPS cell origin. confirming recent studies on the significance of the marrow-derived cells in the intrathymic induction of tolerance ( 10-I 3). Life-long persistence of the tolerant state has long been thought to be ensured by this long persistence of antigen supply, based on stable chimetism ( 14-l 7). Although a correlation between the degree of tolerance and chimetism of thymocytes has been reported ( I71, thymocytes usually MHC-class 11-negative are considered “not” to be involved in tolerance induction. especially for the helper/inducer T cell population (9. 10. 13). Thus, the correlation between tolerance and T cell chimerism may not be the cause, but it is probably the result of induced tolerance. A greater tolerance would always be consistent with longer persistence of the self-renewal compartment of donor cells including tolerance-inducing cells, and it has been experimentally demonstrated that treatments for donor cell destruction promote breakdown of the once-inductul tolerant state ( 16). In the routinely and widely employed experimental system of neonatal tolerance. cells with HPS activity, such as bone marrow (BM) cells and spleen 41s or their mixture. have usually been used. Although the system is practically useful, probably because of the continuous supply of cells participating in tolerance induction, it is rather disadvantageous for cell trace analysis in tolerance induction in the thymus unless the recruits of tolerance-inducing cells from the HPS pool are cut. In this respect. we have reported in a previous paper of this series that peritoneal cavity (PerC) cells and thymus cells are different from BM cells in their cell dose profiles for intrathymic induction and in their duration of’ maintenance of tolerance (9). PerC cells showed more linear compatibility between tolerance inducibility and inoculated cell number than BM cells and only a transient phase of unresponsiveness (for a few weeks after birth). The transient tolerant state in the thymus of newborn mice inoculated with PerC cells was prolonged by the later injection of BM cells without chimerism in thymic lymphocytes (9). Therefore, ii is possible that the transient tolerance after Pert? cell injection is due to a less continuous supply of tolerance-inducing cells. This led us to investigate the cell types and microanatomical sites in the thymus, where negative selection is going on. an&to explore the significance of thymic chimer&m in the induction and maintenance of tolerance. 41A ITR1.4L.S AND METHODS .Wice. BALB/cSgCrSlc( H-T”‘, MIS- 1I’. Thy- 1.I!), DBA/2SgCrSlc(H-2d, Mls- I”., Thy1.2). and (C57BL/h X DBh/:!)F:,(H-2”““. MIS-I h/a. ‘Thy- 1.2: BDF 1) mice were purchased from Japan SLC. Inc. (Shizuoka, Japan). Original AKR/J strain mice (H-Z”, Mls-I”, Thy-l. 1) were purchased fr-om Clea Japan (Tokyo). (BALB/c :i DBA/Z)F, (CDF, 1 and fBALB/c X AKR)F, (CAKF,I mice were produced in our animal facility.
Ia+ CELL CHIMERISM
IN THE TOLERIZED
THYMUS
375
0.5 1 5 10 A mg of lymph node swelling FIG. I. Correlation of V,6+ cell contents in lymphoid cells with their anti-Ml+ I” activities in the local graft-versus-host reaction. Nine samples of mesenteric lymph node cell suspensions containing various amounts of V,6+ cells were obtained from Mls-la-toletized, ?-month-old BALB/c mice. These mice had received intrathymic injections of MB-la-bearing (BALB/c X DBA/2)F, cells, within 24 hr after birth, and complete thymectomy at 1 week ofage. These mice showed different numbers of V,6+ cells in the peripheral lymphocytes and were individually tested in the local graft-versus-host reactions for anti-Mls-la activities (0) anti-alloMHC activities (0) and anti-allo-MHC plus MIS-I” activities (0) in (BALB/c X DBA/Z)F, (H-2d, Mls-1”’ “). (BALB/c X BIO.BR)F, (H-2 d/k, Mls-I”), and (BALB/c X AKR)F, (H-2 “’ , Mls-1 ‘Ia) mice, respectively. Values of V,6+ cell percentage in the lymph node cells were individually plotted against values of PLNswelling activity (Amg of lymph node swelling, Amg = LN weight of cell-injected mice - LN weight of medium-injected mice). r = 0.88 in the response to MIS-I” antigens(p), 0.06 to H-2k antigens (----), and 0.73 to H-2’ plus MIS- la antigens (- . -).
Cell preparation and cell fractionation. Cell suspensions from the thymus and bone marrow (BM) were prepared in Eagle’s minimum essential medium (MEM, Nissui Seiyaku Co., Tokyo) without serum supplement, as described previously (7). PerC cells were collected by washing the peritoneal cavity with 10 ml MEM containing 10% normal syngeneic mouse serum, following anesthetization by an intravenous injection of sodium pentobarbital (Somnopentil, Pitman-Moore, Inc., NJ) (9). Peripheral blood was collected from the retroorbital plexus with a heparinized glass capillary, overlayed on Ficoll-Paque solution (Pharmacia, Uppsala, Sweden). After centrifugation the white blood cells were collected and washed three times in MEM. Antibodies. Mouse mAb to I-Ak (IgG2b, code No. 040-421) and H-2Kk (IgG2a, code No. 030-300) were obtained from Meiji Seiyaku (Tokyo), mAb to Thy-l.2 (30H12, IgG2b), CD4 (GK1.5) from the American Type Culture Collection (Rockville, MD); mAb to the V,6 chain of T cell receptor (44-22- 1, rat IgG2a) was kindly donated by Dr. Hengartner ( 18). These mAb were used for the depletion and staining of cells. FlTC-coupled and uncoupled anti-mouse Ig sheep IgG F(ab); was obtained from Silenus Lab. PTY Ltd. (Victoria, Australia), FITC-sheep anti-rat IgG2 was from Serotec (AAR 04F, Kidtington, Oxford, England), PE-coupled anti-rat-IgG2 was from Biomeda (Foster City, CA), and FITC-coupled peanut agglutinin (PNA) was from Vector Lab. (Burlingame, CA). Tolerance induction. In general, newborn BALB/c (~24 hr old) mice were injected via the anterior facial vein with 50 ~1 MEM containing 2 X lo7 BM cells or 5 X lo6 PerC cells from F, hybrid mice, the cell number showing the minimum dose to induce a maximum level of tolerance in the thymus when assessed 1 week after birth, as described previously (7-9). Some littermates or age-matched neonates that were not in.jected served as positive controls.
376 Tolerized
Normal
i I
I ,
$1 E z -1 7i 0
-
V,6
-
PNA
-
-
FIG. 2. Disappearance vt. V&bearing 1 ~~11s tn the thqmus of young adult BALB/c mouse injected neonatally with Mls- 1“-bone marrow cells. V,6-TcR expression (-) on thymus cells of H-week-old BALB/ c mice (MIS-lb) injected as neonates with 2 Y IO’ BM cells from (BALB/c X AKR)F, mice (Ml~-l”‘~) (right) or not injected (left) was analyzed by fluorocytometry. Thinner lines indicate results without the primary anti-V,,6 mAh. Note that strongly Vi&positive cells (shaded area) were deleted in thymuses tolerized by the iv administration of MIS-I”-BM cells. The deletion is of mature thymocytes or weakly PNR-binding cells (a). but not of strongly PNA-binding or weakly V,6-positive cells (h).
Reconstitution (?f’!l~ ,q~j&i thymes. Six-week-old BALB/c mice were thymectomized as reported previously (19). Four weeks later, they were exposed to lethal, whole-body irradiation (850 rad), and transfused with syngeneic bone marrow cells
Disappearance
of Vi,6 * Cells from ----__-l_-_--___l--~
PBL of BALB/c
Mice
rolerized
Neonatafly
Age of mice (months) --___ State of BALB/c mice --.___ -...Normal Tolerized
to MIS- I ,’
-...-..
.--
.’ .~ _ ..__-.-_-.~
8.26 f 0.40 (II Pi 0.50 +~ 0.20 (fl -- ?I
._..
._ _----
---
4
.-_. _-
8.43 .c I.33 (II 1) 13.75 t_ 0.35 (if 7 4)
to MIS- I ’ Antigens --.--_--.--...------
_... --.--.--.._
I2 ..-.._..
7.61 f 0.44 (II Cf.7 I z 0.16 (n
4) 31
Ia+ CELL CHIMERISM Anti-V 86
IN THE TOLERIZED Anti-Thy.
377
THYMUS
I. I
Anti-Thy.
I .2
BALWC
(H-2d, Mlsb, Thy I .2)
BALE/c. NW. tolerance to(t3ALB,cXDBA/2)F,
/ 1
< 1%
N.T.
1 m
FIG. 3. Disappearance of VB6-bearing T cells in peripheral blood of young adult BALB/c mice tolerized neonatally with Mls-1” bone marrow cells. Proportions and fluorocytometric profiles of V,6-bearing cells and either Thy- 1.l- or Thy- 1.2-bearing cells in the peripheral blood lymphocytes of 2-month-old mice are shown. Newborn BALB/c (Thy-l .2, Mls-lb) mice were injected iv with bone marrow cells from either (BALB/c X DBA/Z)F, (Thy- 1.2, Mls-lb’“) or (BALB/c X AKR)Fi (Thy-l. 1, Mls-1 ‘Ia) mice. N.T., not tested.
(2 X 10’) for reconstitution of the hematolymphoid system. Four weeks later newborn thymus lobes of Mls- 1a’b-semiallogeneic (BALB/c X DBA/2)F, mice were implanted under the kidney capsule. After 4 weeks, the grafted thymus lobes were taken out and the regenerated thymocytes were tested for their immune activity by local graft-versushost reactions. All of the thymus graft recipients were checked for complete removal of their own thymuses. Local host-versus-graft reaction and popliteal lymph node-swelling assay. To assess an ti-Mls- 1a activity of BALB/c thymocytes, the popliteal lymph node (PLN)-swelling assay for local graft-versus-host reaction (GVHR) was employed, as reported previously (7, 20, 21). Briefly, 25 ~1 of MEM containing 2 X 10’ thymus cells was injected into each footpad of adult CDF, and CAKFl hybrid mice. Cells prepared from untreated BALB/c mice and medium only were used as positive controls and negative controls, respectively. Seven days after the footpad injection, the hybrid recipients were killed and PLNs were carefully removed and weighed to within 20.0 1 mg. Antigen-specific responses against allogeneic cellular antigens have been demonstrated in the local GVH assay (7,8). We show here significance of the applied assay system for functional
analyses of V,,6’. MIS-I”-specific -1 cdls (Fig. I). Ihe degree of V,O’ cell content m lymphoid cells seem to reflect directly on the degree of PLN swelling in response to Mls-1” antigens (I’ =r 0.88) but not to allogeneic MHC antigens (I’ z-10.06). The per+ centage of tolerance was calculated as follows: Percentage tolerance =z { 1 - (mean PLN weight of tolerant group - mean PLN weight of medium control)/(mean PLN weight of positive control -- mean PLN weight of medium control) j X 100. Immune activity was also expressed as LN index after dividing PLN weight of the experimental group by that of the medium-injected negative control group (7). Imrlzunoc~?(!flzrorornctric~ ur~l.vsis. C’ell surface antigens were analyzed by flow c! tofluorometrj _as described previously t 9). Aliquots of cell suspensions after treatment of cells with appropriate antibodies conjugated with fluorescein dyes were analyzed by a flow cytometer. FACScan (Recton-Dickinson, Franklin Lakes, NJ). ffi.stohgicu/ o/~,vcn~~~iir)n. Thymuses were frozen immediately after removal and stored in a deepfreezer ( 80°C) until cryostat sections were cut. Frozen sections were air-dried and fixed by acetone for 5 rmn. Immunohistological observations were carried out as follows: for Thy-I--bearing cell determination. cryostat sections were washed with phosphate-buflered saline (PBS) and incubated with one-tenth diluted FITC-antiThy- I mAb. For MHC antigen-bearing cells. slides were incubated with PBS including
la+ CELL CHIMERISM
IN THE TOLERIZED
Normal
THYMUS
379
Tolerized I,
-
V,6
-
FIG. 5. Deletion of VB6-bearing mature T cells in the thymus of l-week-old BALB/c mice injected iv with peritoneal cavity (PerC) cells of MIS- I” mice. One-week-old BALB/c thymus cells were analyzed fluorocytometrically as shown in Fig. 2. Note that strong V,6-positive cells (shaded area) were deleted in the thymus tolerized by iv administration of Mls-la-PC cells.
normal sheep serum (Silenus Lab., Victoria, Australia) and then with anti-mouse Ig sheep Ig F(ab); for 1 hr, for masking of the mouse Ig antigenic determinants. This treatment prevents nonspecific or antigen-cross bindings, if any, of the immune reagents used. The tissue sections were further incubated with either anti-H-2Kk or anti-I-Ak mouse mAb for 40 min at room temperature. After a washing with PBS, FITC-labeled anti-mouse Ig sheep Ig F(ab>‘* solution was dropped on each tissue section. One hour later, specimens were examined by fluorescence microscopy. Statistical analysis. The significance between experimental groups was determined by the two-tailed Mann-Whitney U test. Experimental groups with probability values larger than 0.05 were considered to be not significantly different from the control. RESULTS Deletion .for 1 Year of V&-Bearing T Cells from BALB/c Mice Neonatally Tolerized with Mls-1” Bone Marrow Cells An injection of Mls-l”lb bone marrow (BM) cells from (BALB/c X AKR)F, mice into newborn BALB/c mice (Mls-lb) resulted in hyporesponsiveness. At the age of 2 months, these mice lacked Mls-1” reactive V,6+ cells in the thymus (Fig. 2) showing clonal deletion as the mechanism, as suggested, in our previous paper (8). Immunofluorescence profiles of thymocytes for PNA-binding cells and V,6 TcR-bearing cells were similar in BM cell-injected and uninjected mice, except for deletion of weakly PNA-binding and strongly VB6-positive cells in the BM cell-injected group (Fig. 2). The same was true with TcR gene usage in peripheral T cells. Thus, in contrast to the approximately 10% of peripheral blood lymphocytes (PBL) which are Vp6-bearing T cells in normal mice through 1 year of age, less than 1% of PBL in the tolerized mice were V,6+ cells (Table 1). Thy- 1-allotype marker staining revealed that a low percentage of PBL was donor-derived cells, chimerism being actually present in these mice (Fig. 3). Donor Cell Distribution in the Neonatal Thymus Tolerized with Bone Marrow Cells A few days after F, BM cell injection, BM cell-derived, MHC class I(H-2Kk) and class II(IAk) antigen-bearing cells first appeared in the thymus. These cells were irregular
FIG. 6. lntrathymie distribution ~fdonor-type cells in mice transfused neunatallq peritoneal cavity crtls. Eaperimentai protocols are the same as 111Fig. 3. except given to each neonate. (A, ‘Two-daywld thvmus treated with FI’I‘C-secondaq control. (B) ‘Two-dapold thynus statned for I-A’ antigens. Packed donor-derived and doublets (arrowheads) are evident just outside the vessel. CC) Deep cortex day-old thymus stained for 1-I-31<’ antigens. (Dj C‘ortes OF ?-day-old thymus Magnification. >\13(1.
with Mls-1,’ sem~allogenei~~ that ,? ‘c i (J(’ Per(‘ cells wt:ri antibody only For negati\c cells are seen in vessels. (C’) and medulla (Ml Al’ I”stained for H-7KL antigens
in shape and observed mainly in the medulla with a scattering in the cortex (Fig. 4A!. On Day 7 after BM cell administration, I-/-Zk-positive cells in the cortex often formed clusters of a few cells and a scattering of single cells. This pattern persisted throughout the observation period (Fig. 4BI. On the other hand, in the medulla. I-A’-positive as well as H-2Kk-positive cells were frequently observed as early as 4 days after the F, cell transfusion, and this pattern of donor cell distribution continued throughout the observation period. Thy- 1-bearing cells of the donor’s type first appeared surrounding vessels of the inner cortex and medulla 1 week after MB cell injection. These round lymphocytes formed clusters of a few cells, probably representing an early stage of differentiation of T-lymphocytes in the normally developing thymus (data not shown). Within 1 week, Thy- I ’ cells became single cells scattered throughout the cortex and medulla. It should be noted that in the thymuses of 5-week-old mice. Thy- 1.1 + celts were no longer distributed uniformly. Thus, the Thy- 1.1 + cell-localizing area in the cortex was separated by a clear-cut border line from the area containing no donor’s
Ia+ CELL CHIMERISM
IN THE TOLERIZED
THYMUS
381
FIG. 7. Transient appearance of donor-type cells in the newborn thymus after transfusion with semiallogeneic PerC cells. Donor-type class II-bearing cells, which had once spread through the cortex and medulla on Day 4 (Fig. 6). have started to disappear by I week after the PerC cell injection. Donor cells are still stained in one area (A), but not seen in another area (B) of the cortex in the same cryostat section of the thymus. Magnification, X260.
type T cells; this pattern was not observed in the medulla (summarized later in a chart, Fig. 8). Disappearance of VP6 TcR-Bearing Cells in the Thymus after the Neonatal Administration of Mls-1” Peritoneal Cavity Cells As mentioned above, the second type of tolerance-inducing cell which induces only a transient hyoporesponsiveness, is the PerC cell. Thymus cells of l-week-old BALB/ c mice treated at birth with an injection of Mls-la-PerC cells of (BALB/c X AKR)F, mice lacked Mls-la-reactive, VB6-bearing cells, as shown in Fig. 5. On the other hand, the content of VB8-bearing mature T cells was not affected by this treatment (data not shown). This indicates that clonal deletion is the mechanism of thymus cell tolerance induced by PerC cell injection. Donor Cell Distribution
in the Neonatal Thymus Tolerized by PerC Cell Inoculum
MHC-class I and class II antigen-bearing cells appeared in clusters around vessels in the inner cortex and/or corticomedulla border within a day after the injection of PerC cells (Fig. 6). Within 2 more days, the distribution pattern of these Per-C-derived cells was similar to that observed in the thymus of mice transfused with BM cells 1 week earlier. Thus, PerC cell inoculation created thymus chimer&m of MHC-class IIbearing cells earlier than did BM cells. Morphologically, these cells were not distingushable from those derived from the BM cell inoculum. In a strong contrast to the case after BM cell transfusions, donor type I-Ak-bearing cells began to disappear in the cortex 1 week and in the medulla 2 weeks after the PerC cell transfusion (Fig. 7).
rlcxx3rvo t7 ,\I
382
Days after birth
Cells Transfused into Neonates _-....... ._ .-~.-.-__..-.-.--_-___---.____------l_l-.Bone marrow antiThy-l.1
anti-H-2k (-D.-~-A)
cells % Tolerance
$~;jzl;~;el
Peritoneal anti-
anti-H-$
cells % Tolerance
Thy-l.1
2
N.D. N.D.
7
84 f 6 OiZO
90 -) 4 l&O
49 I_-- f 24 N.D.
21
a5 f a 12 * 2
12 * 9 -.__ o*o
35
N.D. 21 f 5
j&. N.D.
14
FIG. 8. Chart summarizing relationships between the degree of tolerance and tbc intrathvmic distributton pattern of donor-derived cells. Circles with thick and smooth rims and those with thick and rough rims indicate donor-derived T cells and non-f cells, respectively. Some of the data on the degree of tolerance and percentage ofdonor-derived T cells have been reported elsewhere (9). Cell-injected neonates were usually divided into two groups and tested for functional assay and histological analysts. In some groups of Per? cell-treated mice. one lobe of the thymus was used for the assay and the other far the analyses. There was no appreciable differences between H-2Dk-bearing and I-Ak-bearing donor cells in their distribution pattern.
Donor type Thy-I cells were not observed in the thymus after PerC cell injection. All data are summarized later in a chart (Fig. 8). The earlier and more transient appearance of donor type class I- and H-positive celis in the thymus after PerC cell than after MB cell inoculation suggests that BM cells contain precursors for a continuous supply of cells emigrating to the thymus, while PerC cells contain only
la+ CELL CHIMERISM
L
IN THE TOLERIZED
THYMUS
383
I I IItII 5 x) 20 No. of cells/ thymus(x107) I
FIG. 9. Failure of MIS-I” F, thymus epithelium to induce consistent tolerance to MIS determinants in regenerating thymocytes. Thymuses of newborn (BALB/c X DBA/Z) F, mice (MB- lbla, H-2d) were transplanted under the kidney capsule of thymectomized, X-irradiated, syngeneic bone marrow cell-reconstituted BALB/ c (Mls- lb, H-2d) mice. Four weeks later, three thymus grafts (No. 1 - No. 3) were removed and cell suspensions were prepared. Immune activities of the regenerated cells were assessedroutinely by local graft-vs-host assay against the corresponding (BALB/c X DBA/Z)F, antigens (0) and third-party antigens of (C57BL/6 X DBA/ 2)F, (MIS-lb/“, H-2b’d) antigens (0). Horizontal shaded columns indicate the PLN background level (mean 2 SE) in both groups of recipients (n = 5, each).
immediate precursors or immature thymus. Relationships between Donor-Cell
cells with a half life of only a few weeks in the
Distribution
and Tolerant State in the Thymus
The kinetic changes of donor-derived cell distribution patterns in the thymus and the kinetic profiles of intrathymic tolerance were compared and summarized in a chart (Fig. 8). In BM cell groups, tolerance was induced by 1 week after birth, and the unresponsive state with thymus chimer&m lasted throughout the period of observation (3 months after birth). A state of V,6+ cell deletion lasted for a year, as mentioned above. The onset and maintenance of tolerance seem to correlate well with the appearance and presence of donor type MHC-positive cells in the thymus. In Pet-C cell groups the close relationship between the presence/absence of donor type MHC-bearing cells and the degree of tolerance in the thymus is clearly shown. Thus, the tolerance present 1 week after birth starts to disappear during the following week along with the disappearance of donor-derived cells from both medulla and cortex. Thymic Epithelium
May Not Be Suficiently
Involved in Natural Tolerance
Mls- la-semiallogeneic thymuses of newborn CDF,(Mls- 1b’a,H-2d) mice were grafted under the kidney capsule of thymectomized, X-irradiated, BM cell-reconstituted BALB/c mice. Four weeks later, when the generation of host-derived thymocytes would be completed (22), the thymus grafts were removed and the regenerating cells in the thymus grafts were tested for immune activity against the corresponding CDF, antigens and unrelated, third-party antigens of (C57BL/6 X DBA/2)F, mice (Mls-lbia, H-2b’d). The results are shown in Fig. 9. T cells of all three thymus grafts responded equally
well to the third-party antigens, while responses to the corresponding, graft-type antigens fluctuated markedly from hyporesponsiveness to normal responsiveness. It is of interest that more developed thymus grafts contain cells with stronger anti-MIS- 1A activity. so that T cells of thymus grafts of normal physiological size (-7 X 10’ cells/thymus) were not tolerized to MIS- 1a antigens. On the other hand, in the naturally developing thymus V,,6 +* MIS-la-reactive cells appeared within a few days after birth even in Mls1” (BALB/c x AKR)F, mice (Table 2). These autoreactive thymus cells were eliminated within a further few days. Since MHC-class II antigens were continuously expressed on the thymus epithelial cells throughout the observation period (data not shown). elimination of the autoreactive cells may be mediated by factors other than those derived from epithelium. These results indicate that thymus epithelium may lack some factors required for negative selection and is not enough for the consistent induction and maintenance of MIS tolerance l>ISCIJSSION In life-long antigen-specific immunological unresponsiveness, including self-toler,ante, the persistence of tolerance-inducing antigens in a significant concentration has been considered to be a minimum requisite (9, 14, 16. 17). Thus, year-long transplantation tolerance, experimentally induced by neonatal administration of semiallogeneic hemopoietic stem (HPS) cells. is always accompanied by cellular chimerism (9. 14- 17). The significance of hemopoietic cell chimerism for the maintenance of skin-graft tolerance was first demonstrated by Silvers and his colleagues who could terminate the tolerant state after the removal of donor cells by treating tolerized hosts with either immune serum or immune lymphocytes ( 16) capable of attaking, killing, and wiping out the donor cells. although the mechanisms involved in the tolerant state are complex and not yet clarified.
I ransicnl
Group I
i.
II Comparison the groups
of \‘,,6’ .Autoreactw Cells rn the Neonatal Thymus of Mls- I a (BALR/c xi AKR)F, Mice” ~. ~~~.. .-.-_-.~--.-. ~. frradiation rested thymus cells .__. _ ____ I_-_ _..-_------... ( 2000 rad ) rnimediately ige 111 days ‘i V,,h I,?) ‘c i :N (ill heforc GVHR ..-.~..-~_--_-_---.--‘- t rr.ix * li,l71J) i .0X _t 0.2Y (4) Appearance
between (P)
t?I IIi
- () is (.lj i’
i!.O.:
ii.02
* r1.oti (J! X.X.
4utoractivrty tn GVHR: L.N index _f SE t~r(j 1~43 + 0.19 (8) 1.oo _; 0.22 (2) ! 70 + 0.02 (3) i. *. O.lii
I’ Thymus cells individually prepared from (BALB/c X 4KR)F, neonates were assayed for their V,h’. a, well as V,8’ cell contents and for autoreactivity in the syngeneic GVHR. A part of the group I cells were lethally (2000 tad) irradiated immediately before the GVH assay, resulted in 90 % reduction of the activity (I .76 to 0.181 mg of lymph node swelling). This may exclude a possible concomitant reaction of syogeneic host-versus-graft activity. The degree of autoreactivity in group 1 was comparable to the degree of GV’H activity of age-matched BALB/c neonates against (BAL.B/c > AKR)F, antigens. LN index ofthe latter being 2.1 I t 0. I3 (II - 3). Mean 1.N weight of negative control or the medium-injected group was I .55 t 0.3 I mg. N.S., nonsignificant.
Ia+ CELL
CHIMERISM
IN
THE
TOLERIZED
THYMUS
385
Recent findings of restricted usages of TcR V, genes for particular cellular antigens, such as V, 17a to I-E and V,6, 7, 8. I, and 9 to MIS- 1a determinants, and of abrogation of T cells potentially reactive to the corresponding antigenic determinants in the developing thymus as well as in the periphery (23-25) have demonstrated that clonal deletion is a mechanism of self-tolerance. This naturally occurring event is consistently observed in the neonatal tolerance induced in newborn mice by an inoculation of semiallogeneic HPS cells (12). We have here confirmed this phenomenon by using two anatomically different types of cells, bone marrow (BM) cells and peritoneal cavity (PerC) cells. Employing MIS- 1a antigenic determinants and targeting the corresponding V,<6+ T cells, we could demonstrate the tolerance induction mechanism of clonal deletion. The significance of donor cell chimerism in the thymus of neonatally tolerized animals has been documented by Morrissey et al. (17) who analyzed neonatally tolerized mice for the degree of intrathymic chimerism by MHC-class I-bearing cells and the mixed lymphocyte reaction activity and showed a good correlation between the two parameters. Thymocytes with few or no detectable donor cells were partially or fully reactive. Thymuses with partial reactivity to the donor’s antigen were probably in a phase of gradual termination of tolerance, rather than a stable state of “partial tolerance,” since thymocytes of younger tolerant mice were consistently unresponsive in Morrissey’s experiments. Animals with longer lasting hemopoetic activity of donor ceils would have a longer lasting tolerant state, and differences in the degree of the unresponsive state as neonates would lead to differences in the extent of colonization and expansion of semiallogeneic HPS cell compartments in the host. Therefore, an important question is which types of cells are more closely involved in tolerance induction? This will be discussed later. For microanatomical analyses of the donor cell distribution in the thymus in relation to tolerance, one of the minimum requirements is cutting of serial recruitment of semiallogeneic cells entering the thymus. For this, we used two types of Mls-allogeneic inoculum cells: BM cells, a typical compartment with rich HPS activity resulting in year-long tolerance, and PerC cells which can induce intrathymic tolerance within only a few weeks after birth. The transient induction of tolerance by PerC cells is probably due to a lack or shortage of self-renewal of cellular compartments participating in tolerance induction. Indeed, the tolerant state lasted longer when followed by BM cell injection at I week of age (9) probably by the constitution of stable chimerism in the thymus with a continuous supply of tolerance-inducing cells. The long-lasting presence of Ia-bearing cells in the thymus in the BM cell group and their transient appearance in the PerC cell groups also correlated well with the immune state of the organ. The scattered presence in the cortex and the compact presence in the medulla of donor-derived cells showed no difference between the two groups, as long as tolerance was achieved. The gradual disappearance of these cells in the PerC group, first from the cortex then from the medulla, within the first few weeks of life, parallels the gradual disappearance of the tolerant state (9, Fig. 8). Half-lives of the Iaf cells do not show much difference from those after the intrathymic administration of PerC-cell inoculum (M.H.. unpublished observation). It is of interest that the intrathymic half-life of a few weeks of these class II-positive cells, which probably participate in tolerance induction, is similar to the half-life of antigen-presenting stromal cells in the thymus (26). These results remind us that dendritic cells, usually strongly immunopotentiating, become tolerance-inducible for developing T cells in the cultured
thymus (27. 2X). Furthermore, thymic dendritic cells are known to be preferentially localized in the medulla and not in the cortex (29, 30). indeed, direct administration of these cells into the newborn thymus caused unresponsiveness (3 1). Although we could not defined the cell types in a mass of donor’s Iat cells in the neonatally tolerized thymus without the availability of surface markers for allogenic dendtitic cells. the involvement of dendritic cells in the negative selection process seems quite likely. Epithelial cells, another component of la’ cells other than HPS-derived cells, seem not to be sufficiently involved in intrathymic tolerance induction to Mls determinants. as judged from the developmental profile of V,6*. autoreactive thymus cells in Mls1a mice (Table 2) and from the results ofT cell-regeneration experiments in the transplanted thymus (Fig. 9). There, HPS-derived stromal cells, bearing Mls- I’-determinants and remaining in the transplanted Mls- 1a newborn thymus can be considered to he effective in tolerance induction only in a limited number of newly generating thymocytes. However. those results are not against the concept that epithelial ceils can function in negative selection by the passive presentation of Mls determinants, products of HPS-derived stromal cells (32). T cells capable of recognizing MIS- i ‘l determinants. strongly V,?6-positive, were ex-, elusively diminished in the tolerized thymus cells, and the deleted cells are weakly PNA-binding (Fig. 2) or medullary-distributed mature T cells (33). Here. together with the distribution pattern of the injected F, cells. one may consider that the medulla could be the site of negative selection. This is consistent with the finding that V,b ’ cells disappear from the medulla of thymuses tolerized to Mls- 1” antigens (34). However. we should keep it in mind that the disappearance of “self-reactive” cells in the medulla may be the result of the process of signaling for clonal deletion which takes place at an immature, 04’ 8 ’ stage of ‘1‘ cell development probably in the cortex and/or corticomedulla border (34-39). 4CKNOWLEDGMEN-1 We thank for their kind cytofluorometry of Immunology and Culture Technology
S
Drs. H. Lfengartncr and R. MacDonald ofthe Ludwig Lnstrtute for C ancer Research. Lausarme. gift of44-22I hyhridoma; Dr. A. S. Gary for critical comments: and Ms. Y’. Tomita for tloa analyses. This work was supported by a grant from the Shimizu Foundation for the Promotion Research 1989. a grant-in-aid for scientific research from the Ministry ofEducation, Science. and by special coordination funds for promoting science and technology of the Science and Agenc)
REFERENCES I. 2. ‘4. 4. 5. b. 7. 8. 9. 10. I 1. 12.
Hunlg, T.. and Bevan. M. .I.. .I !?\!I. (.IL~L/. lS2, 688, I%0 Wagner. H.. Hardt. C‘.. Heeg, H., Rollinghotf, M., and Pfrzenmarcr, K., Vuruw 284, 27X, 1980. Ando. 1.. and Hurme. M.. ;Ch!rrri, 289, 494. 1981 van Boehmer. H.. and Schubiger. K.. F,u~ ./ Irmn~noi 14, 1048. lYX4. Kast. W. M.. DeWaal. L.. P.. and Meliei: C .J. M.. .I E-\p. ;\l&. 160, 1751. 1984. Biliingham. R. E., Brent. t... and Medawar. P. B.. Phil. Trms. R SM. I~mdon B 239, 357. 1956. Hosono, M.. Kina. T.. Hosokawa. T,, and Katsura, Y.. C’e// fvmtunc)/. 103, 1, 1986. Hosono, M.. and Katsura. Y.. (‘e/I. Irnm~& 103, 1 I. 198h. Hosono. M.. Hosokawa. T.. Kina. 7.. and Katsura. I.., C’ell Imt~~r4nol. 108, 161, 1987. Schwartz. D H.. and Doherty. P. <‘.. ‘Tlr~vn21r.\ 6, LOS. 1984. Ishizaka, S. T.. C‘arnaud. C.. and Stutman, 0.. .I. Imrrzunoi. 137, X47. i Ygb. MacDonald. H. B.. Pedrazzini, T., Schneider. R., I.ouis, J. 4.. Zinkernagel. R M., and Hengartner, .I Elq).
L-i..
la+ CELL CHIMERISM 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 3 I. 32. 33. 34. 35. 36. 37. 38. 39.
IN THE TOLERIZED
THYMUS
387
Nakic, B., Springer, O., Kastelan, A., and Mikuska, J., Immunology 18, 107, 1970. Silobrcic. V., Eur. J, Immunol. 1, 3 13, 197 1. Lubaroff, D. M., and Silvers, W. K., J. Immunol. 111, 65, 1973. Morrissey, P. J., Sharrow, S. O., Kohno, Y., Berzofsky, J. A., and Singer, A., Transplantation 40, 68, 1985. Payne, J., Tuber, B. T., Cannon, N. A., Schneider, R., Schilham, M. W., Acha-Obrea, H., MacDonald, H. R., and Hengartner, H., Proc. Natl. Acad. Sci. USA 85, 7695, 1988. Hosono, M., and Cinader, B., Int. Arch. Allergy Appl. Immunol. 54, 289, 1977. Salaman, M. H., Wedderbum, N., Festenstein, H., and Huber, B., Transplantation 16, 29, 1973. Hosono, M., Kina, T., Katsura, Y., and Hosokawa, T., &and. J. Immunol. 28, 217, 1988. Dukor, P., Miller, J. F. A. P., House, W., and Allman, V., Transplantation 3, 639, 1965. Kappler, J., Roehm, N., and Marrack, P., Cell49, 273, 1987. Kappler, J. W., Staerz, U., White, J., and Marrack, P. C., Nature 332, 35, 1988. MacDonald, H. R., Schneider, R., Lees, R. K., Howe, R. C., Acha-Orbea, H., Festenstein, H., Zinkemagel, R. M., and Hengartner, H. Nature 332,40, 1988. Kyewski, B. A., Fathman, C. G., and Rouse, R. V., J. Exp. Med. 163,231, 1986. Matzinger, P., and Guerder, S., Nature 338, 74, 1989. Mazda, 0.. Watanabe, Y., Gyoutoku, J.-I., and Katsura, Y., J. Exp. Med. 173, 539, 1991. Duijvestiji, A. M., and Barclay, A. N., J. Leuk. Biol. 36, 561, 1984. Rouse, R. V., van Ewijk, W., Jones, P. P., and Weissman, I. L., J. Immunol. 122, 2508, 1979. lnaba, M., Inaba, K., Hosono, M., Kumamoto, T., Ishida, T., Muramatsu, S., Masuda, T., and Ikehara. S., J. Exp. Med. 173, 549, 1991. Ramsdell, F., Lantz, T., and Fowlkes, B. J., Science 246, 1038, 1990. Butcher, E. C., and Weissman, I. L., In “Fundamental Immunology” (W. E. Paul, Ed.), pp. 109-127. Raven Press, New York, 1986. Hengartner, H., Odermatt, B., Schneider, R., Schreyer, M., Walle, G., MacDonald, H. R.. and Zinkemagel, R. M., Nature 336, 388, 1988. Kisielow. P., Bluthmann, H., Staerz, U. D., Steinmetz, M., and von Boehmer, H., Nature 333, 742, 1988. Fowlkes, B. J., Schwartz, R. H., and Pardoll, D. M., Nature334, 620, 1988. MacDonald, H. R., Hengartner, H., and Pedrazzini, T., Nature 335, 174, 1988. Finkel, T. H., Cambier. J. C., Kubo, R. T., Born, W. K., Marrack, P., and Kappler. J. W.. Cell 58, 1047, 1989. MacDonald, H. R., and Lees, R. K., Nature 343, 642, 1990.
IMMUNOLOGY
136.373-387
(199
1)
Neonatal Tolerance Induction in the Thymus to MHC-Class II-Associated Antigens IV. Significance of lntrathymic Chimer&m of Blood-Born la+ Cells in MIS Tolerance MASAMICHI HOSONO,* MAFUMI KUROZUMI,~ MUNEO INABA,$ SHIN IDEYAMA,*‘§ MITSURU TOMANA,* JUN-ICHIRO GYOTOKU,* YOSHIMOTO KATSURA,* AND TOMOHIDE HOSOKAWA~~ Departments of *Immunology and of tClinical Laboratory, Chest Disease Research Institute, and $Department of Dermatology, Faculty of Medicine, Kyoto University, Kyoto, Japan; $Department of Pathology, Kansai-Medical University, Moriguchi, Osaka, Japan; and I’Department of Preventive Medicine, Kyoto Prefecture University of Medicine, Kvoto, Japan Received
November
6, 1990;
accepted
April
2, 1991
The significance of thymus cell chimerism in the induction and maintenance of tolerance was investigated. MIS- I b BALB/c mice were neonatally tolerized by the intravenous administration of either bone marrow (BM) cells or peritoneal cavity (PerC) cells from MIS-lb’” (BALB/c X AKR) F, mice. Tolerance was long-lasting in the BM cell group, but transient in the PerC cell group, probably because PerC cells lack hemopoietic stem cells required for a continuous supply of tolerance-inducing cells. The degree of anti-Mls-la responsiveness of these BALB/c thymus cells was correlated with the degree of intrathymic distribution of the inoculated F, cells. The effect of BM cell inoculation, resulting in a year-long deletion of MIS-la-reactive VB6-bearing T cells is in marked contrast to that of PerC cell inoculation which causesonly a transient loss of V,6+ mature thymocytes (for about 1 week after birth). This functional profile of the tolerant state correlates well with the degree and persistence of the intrathymic presence of F, type Ia+ cells. The longlasting presence of donor-derived cells throughout the thymus tissue in the BM cell group is also in marked contrast to the early disappearance of Ia+ cells (within 2-3 weeks) from the cortex and then from the medulla in the PerC cell group, although these Ia’ cells were once spread throughout the thymus tissue 4 days after the tolerance-inducing cell inoculation. Taken together with a failure to induce consistent unresponsiveness to Mls-1” determinants in Mls-lb thymocytes regenerating in MIS-la-thymic epithelial environments, all the above data indicate that intrathymic chimerism caused by hemopoietic stem cell-derived MHC-class II-bearing cells is a requisite for the induction and maintenance of unresponsiveness by means of clonal deletion in experimentally as well as naturally induced tolerance to MIS determinants. o 1991 Academic ores. IIIC.
INTRODUCTION The thymus is a central organ for differentiation, maturation, and antigen-repertoire generation of T cells, especially of the self-major histocompatibility complex (MHC)class II-restricted helper/inducer subsets ( l-5). In the generation of the T cell repertoire of antigen recognition, thymus T cells directed to self-MHC-class II-associated antigens and possessing autoreactive potentiality are thought to be eliminated by a negative 373 0008-8749/91 $3.00 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
374
tli)SOlrt) t- i .iL
selection process before their peripheralization. Since the pioneer studies of Bilhngham, Brent. and Medawar (6). immunological tolerance induced by the administration of semiallogeneic hemopoietic stem (HPS) cells into newborn mice has been widely employed to clarify the fail-safe mechanisms in the immune system. However. the mechanisms involved are not yet completely clear. In order to investigate the celluiar events in the negative seiechon process in the developing thymus, we employed a tolerance induction system, injecting intravenously hematolymphoid cells of Mls-semiallogeneic F, mice into neonates. The tolerance thus induced in the thymus as early as 1 week after birth was specific to the corre-, sponding antigens (7-9) and closely associated with MHC antigens (8). The effective cells for the induction of tolerance are thought to be of HPS cell origin. confirming recent studies on the significance of the marrow-derived cells in the intrathymic induction of tolerance ( 10-I 3). Life-long persistence of the tolerant state has long been thought to be ensured by this long persistence of antigen supply, based on stable chimetism ( 14-l 7). Although a correlation between the degree of tolerance and chimetism of thymocytes has been reported ( I71, thymocytes usually MHC-class 11-negative are considered “not” to be involved in tolerance induction. especially for the helper/inducer T cell population (9. 10. 13). Thus, the correlation between tolerance and T cell chimerism may not be the cause, but it is probably the result of induced tolerance. A greater tolerance would always be consistent with longer persistence of the self-renewal compartment of donor cells including tolerance-inducing cells, and it has been experimentally demonstrated that treatments for donor cell destruction promote breakdown of the once-inductul tolerant state ( 16). In the routinely and widely employed experimental system of neonatal tolerance. cells with HPS activity, such as bone marrow (BM) cells and spleen 41s or their mixture. have usually been used. Although the system is practically useful, probably because of the continuous supply of cells participating in tolerance induction, it is rather disadvantageous for cell trace analysis in tolerance induction in the thymus unless the recruits of tolerance-inducing cells from the HPS pool are cut. In this respect. we have reported in a previous paper of this series that peritoneal cavity (PerC) cells and thymus cells are different from BM cells in their cell dose profiles for intrathymic induction and in their duration of’ maintenance of tolerance (9). PerC cells showed more linear compatibility between tolerance inducibility and inoculated cell number than BM cells and only a transient phase of unresponsiveness (for a few weeks after birth). The transient tolerant state in the thymus of newborn mice inoculated with PerC cells was prolonged by the later injection of BM cells without chimerism in thymic lymphocytes (9). Therefore, ii is possible that the transient tolerance after Pert? cell injection is due to a less continuous supply of tolerance-inducing cells. This led us to investigate the cell types and microanatomical sites in the thymus, where negative selection is going on. an&to explore the significance of thymic chimer&m in the induction and maintenance of tolerance. 41A ITR1.4L.S AND METHODS .Wice. BALB/cSgCrSlc( H-T”‘, MIS- 1I’. Thy- 1.I!), DBA/2SgCrSlc(H-2d, Mls- I”., Thy1.2). and (C57BL/h X DBh/:!)F:,(H-2”““. MIS-I h/a. ‘Thy- 1.2: BDF 1) mice were purchased from Japan SLC. Inc. (Shizuoka, Japan). Original AKR/J strain mice (H-Z”, Mls-I”, Thy-l. 1) were purchased fr-om Clea Japan (Tokyo). (BALB/c :i DBA/Z)F, (CDF, 1 and fBALB/c X AKR)F, (CAKF,I mice were produced in our animal facility.
Ia+ CELL CHIMERISM
IN THE TOLERIZED
THYMUS
375
0.5 1 5 10 A mg of lymph node swelling FIG. I. Correlation of V,6+ cell contents in lymphoid cells with their anti-Ml+ I” activities in the local graft-versus-host reaction. Nine samples of mesenteric lymph node cell suspensions containing various amounts of V,6+ cells were obtained from Mls-la-toletized, ?-month-old BALB/c mice. These mice had received intrathymic injections of MB-la-bearing (BALB/c X DBA/2)F, cells, within 24 hr after birth, and complete thymectomy at 1 week ofage. These mice showed different numbers of V,6+ cells in the peripheral lymphocytes and were individually tested in the local graft-versus-host reactions for anti-Mls-la activities (0) anti-alloMHC activities (0) and anti-allo-MHC plus MIS-I” activities (0) in (BALB/c X DBA/Z)F, (H-2d, Mls-1”’ “). (BALB/c X BIO.BR)F, (H-2 d/k, Mls-I”), and (BALB/c X AKR)F, (H-2 “’ , Mls-1 ‘Ia) mice, respectively. Values of V,6+ cell percentage in the lymph node cells were individually plotted against values of PLNswelling activity (Amg of lymph node swelling, Amg = LN weight of cell-injected mice - LN weight of medium-injected mice). r = 0.88 in the response to MIS-I” antigens(p), 0.06 to H-2k antigens (----), and 0.73 to H-2’ plus MIS- la antigens (- . -).
Cell preparation and cell fractionation. Cell suspensions from the thymus and bone marrow (BM) were prepared in Eagle’s minimum essential medium (MEM, Nissui Seiyaku Co., Tokyo) without serum supplement, as described previously (7). PerC cells were collected by washing the peritoneal cavity with 10 ml MEM containing 10% normal syngeneic mouse serum, following anesthetization by an intravenous injection of sodium pentobarbital (Somnopentil, Pitman-Moore, Inc., NJ) (9). Peripheral blood was collected from the retroorbital plexus with a heparinized glass capillary, overlayed on Ficoll-Paque solution (Pharmacia, Uppsala, Sweden). After centrifugation the white blood cells were collected and washed three times in MEM. Antibodies. Mouse mAb to I-Ak (IgG2b, code No. 040-421) and H-2Kk (IgG2a, code No. 030-300) were obtained from Meiji Seiyaku (Tokyo), mAb to Thy-l.2 (30H12, IgG2b), CD4 (GK1.5) from the American Type Culture Collection (Rockville, MD); mAb to the V,6 chain of T cell receptor (44-22- 1, rat IgG2a) was kindly donated by Dr. Hengartner ( 18). These mAb were used for the depletion and staining of cells. FlTC-coupled and uncoupled anti-mouse Ig sheep IgG F(ab); was obtained from Silenus Lab. PTY Ltd. (Victoria, Australia), FITC-sheep anti-rat IgG2 was from Serotec (AAR 04F, Kidtington, Oxford, England), PE-coupled anti-rat-IgG2 was from Biomeda (Foster City, CA), and FITC-coupled peanut agglutinin (PNA) was from Vector Lab. (Burlingame, CA). Tolerance induction. In general, newborn BALB/c (~24 hr old) mice were injected via the anterior facial vein with 50 ~1 MEM containing 2 X lo7 BM cells or 5 X lo6 PerC cells from F, hybrid mice, the cell number showing the minimum dose to induce a maximum level of tolerance in the thymus when assessed 1 week after birth, as described previously (7-9). Some littermates or age-matched neonates that were not in.jected served as positive controls.
376 Tolerized
Normal
i I
I ,
$1 E z -1 7i 0
-
V,6
-
PNA
-
-
FIG. 2. Disappearance vt. V&bearing 1 ~~11s tn the thqmus of young adult BALB/c mouse injected neonatally with Mls- 1“-bone marrow cells. V,6-TcR expression (-) on thymus cells of H-week-old BALB/ c mice (MIS-lb) injected as neonates with 2 Y IO’ BM cells from (BALB/c X AKR)F, mice (Ml~-l”‘~) (right) or not injected (left) was analyzed by fluorocytometry. Thinner lines indicate results without the primary anti-V,,6 mAh. Note that strongly Vi&positive cells (shaded area) were deleted in thymuses tolerized by the iv administration of MIS-I”-BM cells. The deletion is of mature thymocytes or weakly PNR-binding cells (a). but not of strongly PNA-binding or weakly V,6-positive cells (h).
Reconstitution (?f’!l~ ,q~j&i thymes. Six-week-old BALB/c mice were thymectomized as reported previously (19). Four weeks later, they were exposed to lethal, whole-body irradiation (850 rad), and transfused with syngeneic bone marrow cells
Disappearance
of Vi,6 * Cells from ----__-l_-_--___l--~
PBL of BALB/c
Mice
rolerized
Neonatafly
Age of mice (months) --___ State of BALB/c mice --.___ -...Normal Tolerized
to MIS- I ,’
-...-..
.--
.’ .~ _ ..__-.-_-.~
8.26 f 0.40 (II Pi 0.50 +~ 0.20 (fl -- ?I
._..
._ _----
---
4
.-_. _-
8.43 .c I.33 (II 1) 13.75 t_ 0.35 (if 7 4)
to MIS- I ’ Antigens --.--_--.--...------
_... --.--.--.._
I2 ..-.._..
7.61 f 0.44 (II Cf.7 I z 0.16 (n
4) 31
Ia+ CELL CHIMERISM Anti-V 86
IN THE TOLERIZED Anti-Thy.
377
THYMUS
I. I
Anti-Thy.
I .2
BALWC
(H-2d, Mlsb, Thy I .2)
BALE/c. NW. tolerance to(t3ALB,cXDBA/2)F,
/ 1
< 1%
N.T.
1 m
FIG. 3. Disappearance of VB6-bearing T cells in peripheral blood of young adult BALB/c mice tolerized neonatally with Mls-1” bone marrow cells. Proportions and fluorocytometric profiles of V,6-bearing cells and either Thy- 1.l- or Thy- 1.2-bearing cells in the peripheral blood lymphocytes of 2-month-old mice are shown. Newborn BALB/c (Thy-l .2, Mls-lb) mice were injected iv with bone marrow cells from either (BALB/c X DBA/Z)F, (Thy- 1.2, Mls-lb’“) or (BALB/c X AKR)Fi (Thy-l. 1, Mls-1 ‘Ia) mice. N.T., not tested.
(2 X 10’) for reconstitution of the hematolymphoid system. Four weeks later newborn thymus lobes of Mls- 1a’b-semiallogeneic (BALB/c X DBA/2)F, mice were implanted under the kidney capsule. After 4 weeks, the grafted thymus lobes were taken out and the regenerated thymocytes were tested for their immune activity by local graft-versushost reactions. All of the thymus graft recipients were checked for complete removal of their own thymuses. Local host-versus-graft reaction and popliteal lymph node-swelling assay. To assess an ti-Mls- 1a activity of BALB/c thymocytes, the popliteal lymph node (PLN)-swelling assay for local graft-versus-host reaction (GVHR) was employed, as reported previously (7, 20, 21). Briefly, 25 ~1 of MEM containing 2 X 10’ thymus cells was injected into each footpad of adult CDF, and CAKFl hybrid mice. Cells prepared from untreated BALB/c mice and medium only were used as positive controls and negative controls, respectively. Seven days after the footpad injection, the hybrid recipients were killed and PLNs were carefully removed and weighed to within 20.0 1 mg. Antigen-specific responses against allogeneic cellular antigens have been demonstrated in the local GVH assay (7,8). We show here significance of the applied assay system for functional
analyses of V,,6’. MIS-I”-specific -1 cdls (Fig. I). Ihe degree of V,O’ cell content m lymphoid cells seem to reflect directly on the degree of PLN swelling in response to Mls-1” antigens (I’ =r 0.88) but not to allogeneic MHC antigens (I’ z-10.06). The per+ centage of tolerance was calculated as follows: Percentage tolerance =z { 1 - (mean PLN weight of tolerant group - mean PLN weight of medium control)/(mean PLN weight of positive control -- mean PLN weight of medium control) j X 100. Immune activity was also expressed as LN index after dividing PLN weight of the experimental group by that of the medium-injected negative control group (7). Imrlzunoc~?(!flzrorornctric~ ur~l.vsis. C’ell surface antigens were analyzed by flow c! tofluorometrj _as described previously t 9). Aliquots of cell suspensions after treatment of cells with appropriate antibodies conjugated with fluorescein dyes were analyzed by a flow cytometer. FACScan (Recton-Dickinson, Franklin Lakes, NJ). ffi.stohgicu/ o/~,vcn~~~iir)n. Thymuses were frozen immediately after removal and stored in a deepfreezer ( 80°C) until cryostat sections were cut. Frozen sections were air-dried and fixed by acetone for 5 rmn. Immunohistological observations were carried out as follows: for Thy-I--bearing cell determination. cryostat sections were washed with phosphate-buflered saline (PBS) and incubated with one-tenth diluted FITC-antiThy- I mAb. For MHC antigen-bearing cells. slides were incubated with PBS including
la+ CELL CHIMERISM
IN THE TOLERIZED
Normal
THYMUS
379
Tolerized I,
-
V,6
-
FIG. 5. Deletion of VB6-bearing mature T cells in the thymus of l-week-old BALB/c mice injected iv with peritoneal cavity (PerC) cells of MIS- I” mice. One-week-old BALB/c thymus cells were analyzed fluorocytometrically as shown in Fig. 2. Note that strong V,6-positive cells (shaded area) were deleted in the thymus tolerized by iv administration of Mls-la-PC cells.
normal sheep serum (Silenus Lab., Victoria, Australia) and then with anti-mouse Ig sheep Ig F(ab); for 1 hr, for masking of the mouse Ig antigenic determinants. This treatment prevents nonspecific or antigen-cross bindings, if any, of the immune reagents used. The tissue sections were further incubated with either anti-H-2Kk or anti-I-Ak mouse mAb for 40 min at room temperature. After a washing with PBS, FITC-labeled anti-mouse Ig sheep Ig F(ab>‘* solution was dropped on each tissue section. One hour later, specimens were examined by fluorescence microscopy. Statistical analysis. The significance between experimental groups was determined by the two-tailed Mann-Whitney U test. Experimental groups with probability values larger than 0.05 were considered to be not significantly different from the control. RESULTS Deletion .for 1 Year of V&-Bearing T Cells from BALB/c Mice Neonatally Tolerized with Mls-1” Bone Marrow Cells An injection of Mls-l”lb bone marrow (BM) cells from (BALB/c X AKR)F, mice into newborn BALB/c mice (Mls-lb) resulted in hyporesponsiveness. At the age of 2 months, these mice lacked Mls-1” reactive V,6+ cells in the thymus (Fig. 2) showing clonal deletion as the mechanism, as suggested, in our previous paper (8). Immunofluorescence profiles of thymocytes for PNA-binding cells and V,6 TcR-bearing cells were similar in BM cell-injected and uninjected mice, except for deletion of weakly PNA-binding and strongly VB6-positive cells in the BM cell-injected group (Fig. 2). The same was true with TcR gene usage in peripheral T cells. Thus, in contrast to the approximately 10% of peripheral blood lymphocytes (PBL) which are Vp6-bearing T cells in normal mice through 1 year of age, less than 1% of PBL in the tolerized mice were V,6+ cells (Table 1). Thy- 1-allotype marker staining revealed that a low percentage of PBL was donor-derived cells, chimerism being actually present in these mice (Fig. 3). Donor Cell Distribution in the Neonatal Thymus Tolerized with Bone Marrow Cells A few days after F, BM cell injection, BM cell-derived, MHC class I(H-2Kk) and class II(IAk) antigen-bearing cells first appeared in the thymus. These cells were irregular
FIG. 6. lntrathymie distribution ~fdonor-type cells in mice transfused neunatallq peritoneal cavity crtls. Eaperimentai protocols are the same as 111Fig. 3. except given to each neonate. (A, ‘Two-daywld thvmus treated with FI’I‘C-secondaq control. (B) ‘Two-dapold thynus statned for I-A’ antigens. Packed donor-derived and doublets (arrowheads) are evident just outside the vessel. CC) Deep cortex day-old thymus stained for 1-I-31<’ antigens. (Dj C‘ortes OF ?-day-old thymus Magnification. >\13(1.
with Mls-1,’ sem~allogenei~~ that ,? ‘c i (J(’ Per(‘ cells wt:ri antibody only For negati\c cells are seen in vessels. (C’) and medulla (Ml Al’ I”stained for H-7KL antigens
in shape and observed mainly in the medulla with a scattering in the cortex (Fig. 4A!. On Day 7 after BM cell administration, I-/-Zk-positive cells in the cortex often formed clusters of a few cells and a scattering of single cells. This pattern persisted throughout the observation period (Fig. 4BI. On the other hand, in the medulla. I-A’-positive as well as H-2Kk-positive cells were frequently observed as early as 4 days after the F, cell transfusion, and this pattern of donor cell distribution continued throughout the observation period. Thy- 1-bearing cells of the donor’s type first appeared surrounding vessels of the inner cortex and medulla 1 week after MB cell injection. These round lymphocytes formed clusters of a few cells, probably representing an early stage of differentiation of T-lymphocytes in the normally developing thymus (data not shown). Within 1 week, Thy- I ’ cells became single cells scattered throughout the cortex and medulla. It should be noted that in the thymuses of 5-week-old mice. Thy- 1.1 + celts were no longer distributed uniformly. Thus, the Thy- 1.1 + cell-localizing area in the cortex was separated by a clear-cut border line from the area containing no donor’s
Ia+ CELL CHIMERISM
IN THE TOLERIZED
THYMUS
381
FIG. 7. Transient appearance of donor-type cells in the newborn thymus after transfusion with semiallogeneic PerC cells. Donor-type class II-bearing cells, which had once spread through the cortex and medulla on Day 4 (Fig. 6). have started to disappear by I week after the PerC cell injection. Donor cells are still stained in one area (A), but not seen in another area (B) of the cortex in the same cryostat section of the thymus. Magnification, X260.
type T cells; this pattern was not observed in the medulla (summarized later in a chart, Fig. 8). Disappearance of VP6 TcR-Bearing Cells in the Thymus after the Neonatal Administration of Mls-1” Peritoneal Cavity Cells As mentioned above, the second type of tolerance-inducing cell which induces only a transient hyoporesponsiveness, is the PerC cell. Thymus cells of l-week-old BALB/ c mice treated at birth with an injection of Mls-la-PerC cells of (BALB/c X AKR)F, mice lacked Mls-la-reactive, VB6-bearing cells, as shown in Fig. 5. On the other hand, the content of VB8-bearing mature T cells was not affected by this treatment (data not shown). This indicates that clonal deletion is the mechanism of thymus cell tolerance induced by PerC cell injection. Donor Cell Distribution
in the Neonatal Thymus Tolerized by PerC Cell Inoculum
MHC-class I and class II antigen-bearing cells appeared in clusters around vessels in the inner cortex and/or corticomedulla border within a day after the injection of PerC cells (Fig. 6). Within 2 more days, the distribution pattern of these Per-C-derived cells was similar to that observed in the thymus of mice transfused with BM cells 1 week earlier. Thus, PerC cell inoculation created thymus chimer&m of MHC-class IIbearing cells earlier than did BM cells. Morphologically, these cells were not distingushable from those derived from the BM cell inoculum. In a strong contrast to the case after BM cell transfusions, donor type I-Ak-bearing cells began to disappear in the cortex 1 week and in the medulla 2 weeks after the PerC cell transfusion (Fig. 7).
rlcxx3rvo t7 ,\I
382
Days after birth
Cells Transfused into Neonates _-....... ._ .-~.-.-__..-.-.--_-___---.____------l_l-.Bone marrow antiThy-l.1
anti-H-2k (-D.-~-A)
cells % Tolerance
$~;jzl;~;el
Peritoneal anti-
anti-H-$
cells % Tolerance
Thy-l.1
2
N.D. N.D.
7
84 f 6 OiZO
90 -) 4 l&O
49 I_-- f 24 N.D.
21
a5 f a 12 * 2
12 * 9 -.__ o*o
35
N.D. 21 f 5
j&. N.D.
14
FIG. 8. Chart summarizing relationships between the degree of tolerance and tbc intrathvmic distributton pattern of donor-derived cells. Circles with thick and smooth rims and those with thick and rough rims indicate donor-derived T cells and non-f cells, respectively. Some of the data on the degree of tolerance and percentage ofdonor-derived T cells have been reported elsewhere (9). Cell-injected neonates were usually divided into two groups and tested for functional assay and histological analysts. In some groups of Per? cell-treated mice. one lobe of the thymus was used for the assay and the other far the analyses. There was no appreciable differences between H-2Dk-bearing and I-Ak-bearing donor cells in their distribution pattern.
Donor type Thy-I cells were not observed in the thymus after PerC cell injection. All data are summarized later in a chart (Fig. 8). The earlier and more transient appearance of donor type class I- and H-positive celis in the thymus after PerC cell than after MB cell inoculation suggests that BM cells contain precursors for a continuous supply of cells emigrating to the thymus, while PerC cells contain only
la+ CELL CHIMERISM
L
IN THE TOLERIZED
THYMUS
383
I I IItII 5 x) 20 No. of cells/ thymus(x107) I
FIG. 9. Failure of MIS-I” F, thymus epithelium to induce consistent tolerance to MIS determinants in regenerating thymocytes. Thymuses of newborn (BALB/c X DBA/Z) F, mice (MB- lbla, H-2d) were transplanted under the kidney capsule of thymectomized, X-irradiated, syngeneic bone marrow cell-reconstituted BALB/ c (Mls- lb, H-2d) mice. Four weeks later, three thymus grafts (No. 1 - No. 3) were removed and cell suspensions were prepared. Immune activities of the regenerated cells were assessedroutinely by local graft-vs-host assay against the corresponding (BALB/c X DBA/Z)F, antigens (0) and third-party antigens of (C57BL/6 X DBA/ 2)F, (MIS-lb/“, H-2b’d) antigens (0). Horizontal shaded columns indicate the PLN background level (mean 2 SE) in both groups of recipients (n = 5, each).
immediate precursors or immature thymus. Relationships between Donor-Cell
cells with a half life of only a few weeks in the
Distribution
and Tolerant State in the Thymus
The kinetic changes of donor-derived cell distribution patterns in the thymus and the kinetic profiles of intrathymic tolerance were compared and summarized in a chart (Fig. 8). In BM cell groups, tolerance was induced by 1 week after birth, and the unresponsive state with thymus chimer&m lasted throughout the period of observation (3 months after birth). A state of V,6+ cell deletion lasted for a year, as mentioned above. The onset and maintenance of tolerance seem to correlate well with the appearance and presence of donor type MHC-positive cells in the thymus. In Pet-C cell groups the close relationship between the presence/absence of donor type MHC-bearing cells and the degree of tolerance in the thymus is clearly shown. Thus, the tolerance present 1 week after birth starts to disappear during the following week along with the disappearance of donor-derived cells from both medulla and cortex. Thymic Epithelium
May Not Be Suficiently
Involved in Natural Tolerance
Mls- la-semiallogeneic thymuses of newborn CDF,(Mls- 1b’a,H-2d) mice were grafted under the kidney capsule of thymectomized, X-irradiated, BM cell-reconstituted BALB/c mice. Four weeks later, when the generation of host-derived thymocytes would be completed (22), the thymus grafts were removed and the regenerating cells in the thymus grafts were tested for immune activity against the corresponding CDF, antigens and unrelated, third-party antigens of (C57BL/6 X DBA/2)F, mice (Mls-lbia, H-2b’d). The results are shown in Fig. 9. T cells of all three thymus grafts responded equally
well to the third-party antigens, while responses to the corresponding, graft-type antigens fluctuated markedly from hyporesponsiveness to normal responsiveness. It is of interest that more developed thymus grafts contain cells with stronger anti-MIS- 1A activity. so that T cells of thymus grafts of normal physiological size (-7 X 10’ cells/thymus) were not tolerized to MIS- 1a antigens. On the other hand, in the naturally developing thymus V,,6 +* MIS-la-reactive cells appeared within a few days after birth even in Mls1” (BALB/c x AKR)F, mice (Table 2). These autoreactive thymus cells were eliminated within a further few days. Since MHC-class II antigens were continuously expressed on the thymus epithelial cells throughout the observation period (data not shown). elimination of the autoreactive cells may be mediated by factors other than those derived from epithelium. These results indicate that thymus epithelium may lack some factors required for negative selection and is not enough for the consistent induction and maintenance of MIS tolerance l>ISCIJSSION In life-long antigen-specific immunological unresponsiveness, including self-toler,ante, the persistence of tolerance-inducing antigens in a significant concentration has been considered to be a minimum requisite (9, 14, 16. 17). Thus, year-long transplantation tolerance, experimentally induced by neonatal administration of semiallogeneic hemopoietic stem (HPS) cells. is always accompanied by cellular chimerism (9. 14- 17). The significance of hemopoietic cell chimerism for the maintenance of skin-graft tolerance was first demonstrated by Silvers and his colleagues who could terminate the tolerant state after the removal of donor cells by treating tolerized hosts with either immune serum or immune lymphocytes ( 16) capable of attaking, killing, and wiping out the donor cells. although the mechanisms involved in the tolerant state are complex and not yet clarified.
I ransicnl
Group I
i.
II Comparison the groups
of \‘,,6’ .Autoreactw Cells rn the Neonatal Thymus of Mls- I a (BALR/c xi AKR)F, Mice” ~. ~~~.. .-.-_-.~--.-. ~. frradiation rested thymus cells .__. _ ____ I_-_ _..-_------... ( 2000 rad ) rnimediately ige 111 days ‘i V,,h I,?) ‘c i :N (ill heforc GVHR ..-.~..-~_--_-_---.--‘- t rr.ix * li,l71J) i .0X _t 0.2Y (4) Appearance
between (P)
t?I IIi
- () is (.lj i’
i!.O.:
ii.02
* r1.oti (J! X.X.
4utoractivrty tn GVHR: L.N index _f SE t~r(j 1~43 + 0.19 (8) 1.oo _; 0.22 (2) ! 70 + 0.02 (3) i. *. O.lii
I’ Thymus cells individually prepared from (BALB/c X 4KR)F, neonates were assayed for their V,h’. a, well as V,8’ cell contents and for autoreactivity in the syngeneic GVHR. A part of the group I cells were lethally (2000 tad) irradiated immediately before the GVH assay, resulted in 90 % reduction of the activity (I .76 to 0.181 mg of lymph node swelling). This may exclude a possible concomitant reaction of syogeneic host-versus-graft activity. The degree of autoreactivity in group 1 was comparable to the degree of GV’H activity of age-matched BALB/c neonates against (BAL.B/c > AKR)F, antigens. LN index ofthe latter being 2.1 I t 0. I3 (II - 3). Mean 1.N weight of negative control or the medium-injected group was I .55 t 0.3 I mg. N.S., nonsignificant.
Ia+ CELL
CHIMERISM
IN
THE
TOLERIZED
THYMUS
385
Recent findings of restricted usages of TcR V, genes for particular cellular antigens, such as V, 17a to I-E and V,6, 7, 8. I, and 9 to MIS- 1a determinants, and of abrogation of T cells potentially reactive to the corresponding antigenic determinants in the developing thymus as well as in the periphery (23-25) have demonstrated that clonal deletion is a mechanism of self-tolerance. This naturally occurring event is consistently observed in the neonatal tolerance induced in newborn mice by an inoculation of semiallogeneic HPS cells (12). We have here confirmed this phenomenon by using two anatomically different types of cells, bone marrow (BM) cells and peritoneal cavity (PerC) cells. Employing MIS- 1a antigenic determinants and targeting the corresponding V,<6+ T cells, we could demonstrate the tolerance induction mechanism of clonal deletion. The significance of donor cell chimerism in the thymus of neonatally tolerized animals has been documented by Morrissey et al. (17) who analyzed neonatally tolerized mice for the degree of intrathymic chimerism by MHC-class I-bearing cells and the mixed lymphocyte reaction activity and showed a good correlation between the two parameters. Thymocytes with few or no detectable donor cells were partially or fully reactive. Thymuses with partial reactivity to the donor’s antigen were probably in a phase of gradual termination of tolerance, rather than a stable state of “partial tolerance,” since thymocytes of younger tolerant mice were consistently unresponsive in Morrissey’s experiments. Animals with longer lasting hemopoetic activity of donor ceils would have a longer lasting tolerant state, and differences in the degree of the unresponsive state as neonates would lead to differences in the extent of colonization and expansion of semiallogeneic HPS cell compartments in the host. Therefore, an important question is which types of cells are more closely involved in tolerance induction? This will be discussed later. For microanatomical analyses of the donor cell distribution in the thymus in relation to tolerance, one of the minimum requirements is cutting of serial recruitment of semiallogeneic cells entering the thymus. For this, we used two types of Mls-allogeneic inoculum cells: BM cells, a typical compartment with rich HPS activity resulting in year-long tolerance, and PerC cells which can induce intrathymic tolerance within only a few weeks after birth. The transient induction of tolerance by PerC cells is probably due to a lack or shortage of self-renewal of cellular compartments participating in tolerance induction. Indeed, the tolerant state lasted longer when followed by BM cell injection at I week of age (9) probably by the constitution of stable chimerism in the thymus with a continuous supply of tolerance-inducing cells. The long-lasting presence of Ia-bearing cells in the thymus in the BM cell group and their transient appearance in the PerC cell groups also correlated well with the immune state of the organ. The scattered presence in the cortex and the compact presence in the medulla of donor-derived cells showed no difference between the two groups, as long as tolerance was achieved. The gradual disappearance of these cells in the PerC group, first from the cortex then from the medulla, within the first few weeks of life, parallels the gradual disappearance of the tolerant state (9, Fig. 8). Half-lives of the Iaf cells do not show much difference from those after the intrathymic administration of PerC-cell inoculum (M.H.. unpublished observation). It is of interest that the intrathymic half-life of a few weeks of these class II-positive cells, which probably participate in tolerance induction, is similar to the half-life of antigen-presenting stromal cells in the thymus (26). These results remind us that dendritic cells, usually strongly immunopotentiating, become tolerance-inducible for developing T cells in the cultured
thymus (27. 2X). Furthermore, thymic dendritic cells are known to be preferentially localized in the medulla and not in the cortex (29, 30). indeed, direct administration of these cells into the newborn thymus caused unresponsiveness (3 1). Although we could not defined the cell types in a mass of donor’s Iat cells in the neonatally tolerized thymus without the availability of surface markers for allogenic dendtitic cells. the involvement of dendritic cells in the negative selection process seems quite likely. Epithelial cells, another component of la’ cells other than HPS-derived cells, seem not to be sufficiently involved in intrathymic tolerance induction to Mls determinants. as judged from the developmental profile of V,6*. autoreactive thymus cells in Mls1a mice (Table 2) and from the results ofT cell-regeneration experiments in the transplanted thymus (Fig. 9). There, HPS-derived stromal cells, bearing Mls- I’-determinants and remaining in the transplanted Mls- 1a newborn thymus can be considered to he effective in tolerance induction only in a limited number of newly generating thymocytes. However. those results are not against the concept that epithelial ceils can function in negative selection by the passive presentation of Mls determinants, products of HPS-derived stromal cells (32). T cells capable of recognizing MIS- i ‘l determinants. strongly V,?6-positive, were ex-, elusively diminished in the tolerized thymus cells, and the deleted cells are weakly PNA-binding (Fig. 2) or medullary-distributed mature T cells (33). Here. together with the distribution pattern of the injected F, cells. one may consider that the medulla could be the site of negative selection. This is consistent with the finding that V,b ’ cells disappear from the medulla of thymuses tolerized to Mls- 1” antigens (34). However. we should keep it in mind that the disappearance of “self-reactive” cells in the medulla may be the result of the process of signaling for clonal deletion which takes place at an immature, 04’ 8 ’ stage of ‘1‘ cell development probably in the cortex and/or corticomedulla border (34-39). 4CKNOWLEDGMEN-1 We thank for their kind cytofluorometry of Immunology and Culture Technology
S
Drs. H. Lfengartncr and R. MacDonald ofthe Ludwig Lnstrtute for C ancer Research. Lausarme. gift of44-22I hyhridoma; Dr. A. S. Gary for critical comments: and Ms. Y’. Tomita for tloa analyses. This work was supported by a grant from the Shimizu Foundation for the Promotion Research 1989. a grant-in-aid for scientific research from the Ministry ofEducation, Science. and by special coordination funds for promoting science and technology of the Science and Agenc)
REFERENCES I. 2. ‘4. 4. 5. b. 7. 8. 9. 10. I 1. 12.
Hunlg, T.. and Bevan. M. .I.. .I !?\!I. (.IL~L/. lS2, 688, I%0 Wagner. H.. Hardt. C‘.. Heeg, H., Rollinghotf, M., and Pfrzenmarcr, K., Vuruw 284, 27X, 1980. Ando. 1.. and Hurme. M.. ;Ch!rrri, 289, 494. 1981 van Boehmer. H.. and Schubiger. K.. F,u~ ./ Irmn~noi 14, 1048. lYX4. Kast. W. M.. DeWaal. L.. P.. and Meliei: C .J. M.. .I E-\p. ;\l&. 160, 1751. 1984. Biliingham. R. E., Brent. t... and Medawar. P. B.. Phil. Trms. R SM. I~mdon B 239, 357. 1956. Hosono, M.. Kina. T.. Hosokawa. T,, and Katsura, Y.. C’e// fvmtunc)/. 103, 1, 1986. Hosono, M.. and Katsura. Y.. (‘e/I. Irnm~& 103, 1 I. 198h. Hosono. M.. Hosokawa. T.. Kina. 7.. and Katsura. I.., C’ell Imt~~r4nol. 108, 161, 1987. Schwartz. D H.. and Doherty. P. <‘.. ‘Tlr~vn21r.\ 6, LOS. 1984. Ishizaka, S. T.. C‘arnaud. C.. and Stutman, 0.. .I. Imrrzunoi. 137, X47. i Ygb. MacDonald. H. B.. Pedrazzini, T., Schneider. R., I.ouis, J. 4.. Zinkernagel. R M., and Hengartner, .I Elq).
L-i..
la+ CELL CHIMERISM 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 3 I. 32. 33. 34. 35. 36. 37. 38. 39.
IN THE TOLERIZED
THYMUS
387
Nakic, B., Springer, O., Kastelan, A., and Mikuska, J., Immunology 18, 107, 1970. Silobrcic. V., Eur. J, Immunol. 1, 3 13, 197 1. Lubaroff, D. M., and Silvers, W. K., J. Immunol. 111, 65, 1973. Morrissey, P. J., Sharrow, S. O., Kohno, Y., Berzofsky, J. A., and Singer, A., Transplantation 40, 68, 1985. Payne, J., Tuber, B. T., Cannon, N. A., Schneider, R., Schilham, M. W., Acha-Obrea, H., MacDonald, H. R., and Hengartner, H., Proc. Natl. Acad. Sci. USA 85, 7695, 1988. Hosono, M., and Cinader, B., Int. Arch. Allergy Appl. Immunol. 54, 289, 1977. Salaman, M. H., Wedderbum, N., Festenstein, H., and Huber, B., Transplantation 16, 29, 1973. Hosono, M., Kina, T., Katsura, Y., and Hosokawa, T., &and. J. Immunol. 28, 217, 1988. Dukor, P., Miller, J. F. A. P., House, W., and Allman, V., Transplantation 3, 639, 1965. Kappler, J., Roehm, N., and Marrack, P., Cell49, 273, 1987. Kappler, J. W., Staerz, U., White, J., and Marrack, P. C., Nature 332, 35, 1988. MacDonald, H. R., Schneider, R., Lees, R. K., Howe, R. C., Acha-Orbea, H., Festenstein, H., Zinkemagel, R. M., and Hengartner, H. Nature 332,40, 1988. Kyewski, B. A., Fathman, C. G., and Rouse, R. V., J. Exp. Med. 163,231, 1986. Matzinger, P., and Guerder, S., Nature 338, 74, 1989. Mazda, 0.. Watanabe, Y., Gyoutoku, J.-I., and Katsura, Y., J. Exp. Med. 173, 539, 1991. Duijvestiji, A. M., and Barclay, A. N., J. Leuk. Biol. 36, 561, 1984. Rouse, R. V., van Ewijk, W., Jones, P. P., and Weissman, I. L., J. Immunol. 122, 2508, 1979. lnaba, M., Inaba, K., Hosono, M., Kumamoto, T., Ishida, T., Muramatsu, S., Masuda, T., and Ikehara. S., J. Exp. Med. 173, 549, 1991. Ramsdell, F., Lantz, T., and Fowlkes, B. J., Science 246, 1038, 1990. Butcher, E. C., and Weissman, I. L., In “Fundamental Immunology” (W. E. Paul, Ed.), pp. 109-127. Raven Press, New York, 1986. Hengartner, H., Odermatt, B., Schneider, R., Schreyer, M., Walle, G., MacDonald, H. R.. and Zinkemagel, R. M., Nature 336, 388, 1988. Kisielow. P., Bluthmann, H., Staerz, U. D., Steinmetz, M., and von Boehmer, H., Nature 333, 742, 1988. Fowlkes, B. J., Schwartz, R. H., and Pardoll, D. M., Nature334, 620, 1988. MacDonald, H. R., Hengartner, H., and Pedrazzini, T., Nature 335, 174, 1988. Finkel, T. H., Cambier. J. C., Kubo, R. T., Born, W. K., Marrack, P., and Kappler. J. W.. Cell 58, 1047, 1989. MacDonald, H. R., and Lees, R. K., Nature 343, 642, 1990.