Proc. Natl. Acad. Sci. USAVol. 82, pp. 2483-2487, April 1985Medical Sciences
Rationale for bone marrow transplantation in the treatment ofautoimmune diseases
(bone marrow transplantation/BXSB mice/MRL/I mice/T-cel functions)
SUSUMU IKEHARA*, ROBERT A. GOODt, TAKAO NAKAMURA*, KEN'ICHI SEKITA*, SHUJI INOUE*,MAUNG MAUNG 00*, ERI Muso*, KATSUHIKO OGAWA*, AND YOSHIHIRo HAMASHIMA**Department of Pathology, Faculty of Medicine, Kyoto University, Sakyo-ku, Kyoto 606, Japan; and tOklahoma Medical Research Foundation, 825 Northeast13th Street, Oklahoma City, OK 73104
Contributed by Robert A. Good, December 3, 1984
ABSTRACT Transplantation of normal bone marrowfrom C3H/HeN nu/nu (H-2k) mice into young MRL/MP-lpr/lpr (MRL/I; H-2k) mice (<1.5 mo) prevented the develop-ment of autoimmune diseases and characteristic thymic abnor-malities in the recipient mice. When female MRL/1 (>2 mo)or male BXSB (H-2b) mice (9 mo) with autoimmune diseasesand lymphadenopathy were lethally irradiated and then recon-stituted with allogeneic bone marrow cells from youngBALB/c nu/nu (H-2d) mice (<2 mo), the recipients survivedfor more than 3 mo after the bone marrow transplantation andshowed no graft-versus-host reaction. Histopathological studyrevealed that lymphadenopathy disappeared and that all evi-dence of autoimmune disease either was prevented from devel-oping or was completely corrected even after its developmentin such mice. All abnormal T-cell functions were restored tonormal. The newly developed T cells were found to be tolerantof both bone marrow donor-type (BALB/c) and host-type(MRL/I or BXSB) major histocompatibility complex (MHC)determinants. Therefore, T-cell dysfunction in autoimmune-prone mice can be associated with both the involutionarychanges that occur in the thymus of the autoimmune-pronemice and also to abnormalities that reside in the stem cells.However, normal stem cells from BALB/c nu/nu donors candifferentiate into normal functional T cells even in mice whosethymus had undergone considerable involution, as in the caseof BXSB or MRL/I mice in the present studies. These findingssuggest that marrow transplantation may be a strategy ulti-mately to be considered as an approach to treatment of life-threatening autoimmune diseases in humans. T-cell dysfunc-tion in autoimmune-prone mice previously attributed to invo-lutionary changes that occur in the thymus of these mice mayinstead be attributed to abnormalities that basically reside inthe stem cells of the autoimmune-prone mice.
The availability of several murine strains that develop sys-temic lupus erythematosus-like disease has prompted effortsto gain better understanding of the fundamental nature ofautoimmune diseases through extensive studies of the immu-nological abnormalities of these mice. MRL/MP-lpr/lpr(MRL/l) and BXSB mice as well as NZB mice and NZB xNZW F1 hybrids spontaneously develop autoimmune dis-eases characterized by anti-double-stranded (ds) DNA anti-bodies, immune-complex glomerulonephritis, and deathfrom renal failure (1). Abnormalities have been found in orattributed to T cells, thymic epithelium, B cells, and/or mac-rophages in these mice (2-8). Recently, several groups haveshown that the proneness to develop autoimmune diseasesactually resides in defects at the lymphoid stem-cell leveland that defects of function are not directly attributable toenvironmental factors such as hormones or viruses (9-11).
Therefore, we examined whether or not transfer of normalstem cells into autoimmune-prone mice can be used to bothprevent and treat autoimmune diseases.
In the present study, we show that allogeneic bone mar-row transplantation from donors carrying the nu/nu genespermits both prevention and effective treatment of autoim-mune diseases in both MRL/l and BXSB mice and that nor-mal T cells, B cells, and macrophages developed in suchmice transplanted with allogeneic bone marrow 3 mo afterthe bone marrow transplantation.
MATERIALS AND METHODSMice. Inbred MRL/l, MRL/MP-+/+ (MRL/n), and
BXSB mice were obtained from The Jackson Laboratory.BALB/c nu/nu, BALB/c, C3H/HeN nu/nu, C3H/HeN,and DBA/2 mice were obtained from the Central Institutefor Experimental Animals (Tokyo). These mice were main-tained under specific pathogen-free conditions in our facili-ties.Bone Marrow Transplantation. MRL/l and BXSB mice
were irradiated (850 rads for MRL/l and 950 rads for BXSBfrom a cobalt-60 source) and reconstituted by intravenousinjection of 2 x 107 bone marrow cells from young BALB/cnu/nu mice (<2 mo). These MRL/l and BXSB mice wereusually sacrificed more than three months after bone mar-row transplantation. Testing with anti-H-2 antisera and com-plement indicated that more than 90% of spleen cells were ofdonor type.
Cell Separation. Mice were sacrificed by cervical disloca-tion. The spleens were removed aseptically, minced, andgently passed through a fine mesh stainless-steel sieve intophosphate-buffered saline.
Cytotoxic Test. Spleen cells that were suspended in TC-199(GIBCO) with 5% fetal calf serum (Microbiological Asso-ciates) were adjusted to 5 x 106 cells per ml and divided intotwo aliquots of 50 ,ul. The cells were incubated with 50 ,ul ofa 1:10 dilution of anti-Thy-1.2 antibody for 30 min at 4°C.The cells were then washed once and resuspended in 100 ,ulof a 1:10 dilution of rabbit complement (Cedarlane Labora-tories, Hornby, ON, Canada) previously absorbed withmouse spleen cells. After a 30-min incubation at 37°C, theviability of the cells was determined by trypan blue dye-ex-clusion test. The counts were converted to a cytotoxic indexby the formula:
CI =% V cells (C alone) - % V cells (Ab + C) x 100% V cells (C alone)
in which C represents complement; Ab, antibody; CI, cyto-toxic index; and V cells, viable cells.
Abbreviations: MRL/l, MRL/MP-lpr/lpr; CICs, circulating im-mune complexes; MHC, major histocompatibility complex; Con A,concanavalin A; ds, double-stranded.
2483
The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.
Dow
nloa
ded
by g
uest
on
Oct
ober
1, 2
020
2484 Medical Sciences: Ikehara etaLP
Mitogen Response. The mitogenic reactivity was deter-mined by measuring incorporation of [3H]thymidine intoDNA. Triplicate cultures were set up in wells of flat-bottommicrotiter plates (Corning Glass Works 25860, Corning,NY), each containing 5 x 105 cells in 0.2 ml of RPMI-1640medium (Nissui Seiyaku Co., Tokyo) that was supplementedwith 2 mM L-glutamine (Wako Pure Chemical, Tokyo), peni-cillin (100 units/ml), streptomycin (100 ug/ml), and 5% hu-man plasma. The cells were cultured in the presence of phy-tohemagglutinin P (Difco) at 25 pg/ml or of concanavalin A(Con A; Calbiochem) at 5.0 ,ug/ml. (The mitogens had beentested at various doses, and the dose selected was that givingthe optimal responses.) The cultures were incubated for 72hr at 37°C in a humidified atmosphere of 5% C02/95% air.[3H]Thymidine (0.5 ,Ci in 20 .ul; New England Nuclear; 1 Ci= 37 GBq) was present during the last 4 hr of the culturingperiod. The 3H radioactivity incorporated into trichloroace-tic acid-insoluble material was measured by a liquid scintilla-tion counter.
Generation of Cytotoxic T Lymphocytes. Responder cells(7.5 x 106) and mitomycin C (50 ug/ml)-treated stimulatorcells (2.5 x 106) were cocultured in RPMI-1640 medium con-taining 10% heat-inactivated human serum supplementedwith streptomycin (100 pg/mi), penicillin (100 units/ml) and2-mercaptoethanol (50 tM). After 5 days of coculture in ahumidified CO2 incubator, the cells were collected, and their
cytotoxic activity was determined by 51Cr-release assay asdescribed (12).Measurement of Circulating Immune Complex (CIC). CICs
were measured by using microtiter plates coated with com-plement component Clq as described (13).
Antibodies to dsDNA. Antibodies to dsDNA were mea-sured by solid-phase binding assay as described (14).
RESULTS AND DISCUSSIONAutoimmune-prone mice, such as mice of (NZB x NZW)F1, MRL/l, and BXSB strains, show premature thymic invo-lution, including morphological and functional abnormalities(15-17). In order to prove why thymic abnormalities developin autoimmune-prone mice, transplantation of the thymusand bone marrow was performed. When thymuses of new-born MRL/l (H-2") mice were grafted into C3H/HeN nu/nu(H-2") mice, the thymuses of the recipients did not show pre-mature involution even 5 mo after grafting (0/11). By con-trast, when thymuses of newborn normal C3H/HeN micewere grafted into MRL/l mice, the thymuses showed prema-ture involution (8/9). Thus, premature thymic involution inautoimmune-prone mice is not intrinsically determined with-in the thymus but is secondary to other events. When normalbone marrow cells (2 x 107) of young C3H/HeN nu/nu micewere transferred to irradiated young MRL/l mice (<1.5 mo),
~~~1'-~ ~ ~ ~ ~ 0.,. k,, 9#.
;' "'9
;02,*. 4t.t X* Cv b H%*t /;§ w Nt ' t* z 4- ~~~~~~~~
0AX/sg ;%< l-f
jr > . *w ~~~~ - llj, '; r ff~~I eit *
t~~~~~~~t~~jo7tS ~ -0> 4 i 1 X 9 ' 2 ).ws0, jiA,j5 t;t ;r**5tg*i#*.
,,4 >e.*? S 4; ;+ '
"9-4 ..fetI~e< X ,/ Z#A'
14* .;lrW~o' to, V
s^ ...rJ;n slD ; 1s~ *e e;F,~~,4V~* .~9.
FIG. 1. Histopathological findings of 5-mo-old MRL/l mice reconstituted with bone marrow cells of young BALB/c nu/nu (<2 mo) orMRL/l (<1.5 mo) mice. (A and B) Kidney (A) and liver (B) of MRL/l mouse reconstituted with bone marrow cells of MRL/l mice show amarked infiltration of lymphoid cells into the perivascular areas; particularly to be noted are glomerular sclerosis and protein casts in tubuli. (Cand D) Kidney (C) and liver (D) of MRL/I mouse reconstituted with bone marrow cells of BALB/c nu/nu mice demonstrate only a slightinfiltration of lymphoid cells into the perivascular areas of the kidney and the liver. The appearance of most of the glomeruli is within normalrange.
Proc. NatL Acad ScL USA 82 (1985)
Dow
nloa
ded
by g
uest
on
Oct
ober
1, 2
020
Proc. Natl. Acad Sci. USA 82 (1985) 2485
thymic abnormalities and autoimmune diseases did not de-velop up to 5 mo after the marrow transplant (0/5). It seemsreasonable to conclude from these findings that abnormalstem cells of the autoimmune-prone mice are involved in amajor way in the premature thymic involution and develop-ment of autoimmune diseases in these mice. These findingsare in agreement with published observations of Jyonouchiet al. who showed that transplants within the sgme majorhistocompatibility complex (MHC) corrected fully the im-munologic and hematologic abnormalities that characterizedB/W F. mice (11).The next step was to verify whether or not transfer of nor-
mal stem cells into autoimmune-prone mice could be used totreat autoimmune diseases in these animals. When femaleMRL/l (H-2k) (>2 mo) or male BXSB (H-2b) (>9 mo) micethat had already shown clear evidence of autoimniune dis-eases and lymphadenopathy were lethally irradiated and re-constituted with allogeneic bone marrow cells from youngBALB/c nu/nu (H-2d) donors (<2 mo), the recipients sur-vived in good health for more than 3 mo after bone marrowtransplantation without showing any evidence of graft-ver-sus-host reaction either under specific pathogen-free condi-tions or in a conventional environment. By contrast, whenMRL/l (>2 mo) mice were lethally irradiated and reconsti-tuted with bone marrow cells of MRL/l (<1.5 mo) mice, therecipients regularly died of progressive renal failure within 2mo.
Histopathological study demonstrated that lymphadenop-athy disappeared and that characteristic lymphoid cell infil-trations into the kidney and liver were markedly amelioratedin both MRL/l and BXSB mice that had been reconstitutedwith bone marrow cells of BALB/c nu/nu mice (Fig. 1). Animmunofluorescence study revealed that glomerular depos-its of IgG, IgA, C3, and glycoprotein 70 (gp70) were all mark-edly reduced in the mice reconstituted with BALB/c nu/lnubone marrow cells; in particular, deposits of complement
Table 1. Correlation between circulating immune complex (CIC)levels and glomerular damages
CICs, Anti-dsDNA, GlomerularMice Age,* mo ,ug/ml cpm damagest
BALB/ct >6 <1 437BXSBf 8 143 2,466 + + +MRL/lt 4 102 22,146 + + +BXSB 9 -*11 1 408 +
9 12 <1 1,048 +9 -*12.5 2 948 +9.5 14 9 1,182 +9.5-14 2 1,695 +9.5 15 1 417 +
12 16 34 995 +13.5 16.5 5 1,254 +14 18 6 1,251 +
MRL/l 2 4 <1 570 +2 6 80 35,658 +3 8 2 1,651 +3.5 7 34 30,481 +4 6.5 4 1,253 +
CICs were measured by using microtiter plates coated withcomplement component Clq as described (13). Antibodies todsDNA were measured by solid-phase binding assay (14).*9 -* 11, etc., means the BXSB mouse was irradiated at the age of9 mo (or more as indicated), reconstituted with bone marrow cellsof BALB/c nu/nu mice, and sacrificed at the age of 11 mo (or moreas indicated), and the serum was used for the assays.t+, Minimal glomerular injury.tPooled sera from BALB/c (>6 mo) were used as the negativecontrol, and pooled sera from 8-mo-old BXSB and 4-mo-oldMRL/I mice were used as positive controls.
component C3 were almost completely absent from the kid-neys of these mice.
It is well known that autoimmune-prone mice show highlevels of CICs and anti-dsDNA antibodies with age (18, 19).We measured antibodies to dsDNA and CICs as previouslydescribed. As shown in Table 1, BXSB and MRL/l mice re-constituted with normal bone marrow cells exhibited signifi-cantly lower levels of CICs and anti-dsDNA antibodies incomparison with control BXSB and MRL/l mice. The de-gree of glomerular damagt was correlated with the CIC lev-els in these mice.
Recently, Wofsy et al. (20) have reported that BXSB3 miceshow a progressive increase in the number of peripheralblood monocytes. Following the method they described, wecounted the number of monocytes in the peripheral blood ofBXSB mice reconstituted with normal bone marrow cells ofBALB/c nu/nu mice and compared this number with that ofmonocytes present in blood of untreated BXSB mice. Thenumber of monocytes in the BXSB mice was reduced to thenormal range (<10%), whereas untreated BXSB mice (8 moold) showed a marked monocytosis (>60%) in the peripheralblood. In addition, polyclonal B-cell hyperactivity in MRL/land BXSB mice, which hag been reported (10, 18), also wasreduced to normal levels after reconstitution with normalbone marrow across the MHC barriers.
In contrast, T-cell functions were increased as shown inFig. 2: untreated MRL/l mice at the age of 3 mo exhibited ahigh number of Thy-1.2-positive cells in their spleen but alow responsiveness to Con A. However, MRL/l and BXSBmice reconstituted with normal bone marrow cells showed asignificantly higher responsiveness to Con A than did un-treated controls. Using anti-H-2d, anti-H-2b, and anti-H-2kantisera plus complement, we confirmed that the newly de-veloped T cells in MRL/l and BXSB mice were all derivedfrom stem cells ofBALB/c nu/nu mice. The assay for induc-tion of cytotoxic T lymphocytes revealed that the T cellswere tolerant to both bone marrow donor-type (BALB/c)and host-type (MRL/l or BXSB) MHC determinants (Table2). These results are consistent with findings (21-24) in radi-ation bone marrow chimeras, where MHC barriers werebreached by using nonautoimmune mice. In the untreated
ConA Response (Stimulation Index)
Transplantation Sacrifice 0(mo) (mo)
MRL/l0)(-) 32 - 42 > 6
3.5 - 6.54 - 6.5
3.5 - 7
BXSB (C) (-)9 -9 -9 -
9.5 -9.5 7
13.5 -
8
1 22.5
1 41 56.5
5 10 15 20
- _
-J
0 20 40 60 180 100
% THY- 1.2 Cells
FIG. 2. Restoration of responsiveness to in vitro stimulation ofspleen cells with Con A after bone marrow transplantation fromBALB/c nu/nu donors into MRL/l or BXSB recipients. Numbersof Thy-1.2-positive cells (c) and Con A response (U) are indicated.The mice were reconstituted at the times indicated in the columnunder transplantation and were sacrificed for spleen cell analysis atthe times indicated under sacrifice.
Medical Sciences: Ikehara et aL
0
Dow
nloa
ded
by g
uest
on
Oct
ober
1, 2
020
2486 Medical Sciences: Ikehara etalP
Table 2. Newly-developed T cells are tolerant to both bone marrow (BM) donor-type (BALB/c) and host-type(MRL/l or BXSB) MHC determinants
% specific release from target cells
Age, (mean ± s.d.)mo Mice E/T ratio* P815(H-2d) EL-4(H_2b) X5563(H_2k)
2 BALB/c(H-2d) 2/1 4.2 ± 0.6 79.6 ± 9.4 1002 C3H/HeN(H-2k) 2/1 51.4 ± 16.3 59.7 ± 0.5 03 MRL/l (H-2&) 4/1(NW)t 2.0 ± 0.5 0 1.2 ± 0.76 MRL/l with BALB/c nu/nu BMt 4/1(NW) 0 46.2 ± 0.6 09 BXSB(H_2b) 4/1(NW) 1.4 ± 0.6 0 1.0 ± 0.512 BXSB with BALB/c nu/nu BM 4/1(NW) 0 0.5 ± 2.2 85.7 ± 3.4
Responder cells (7.5 x 106) and mitomycin C (50 jug/ml)-treated stimulator cells (2.5 x 101) were cocultured inRPMI-1640 medium containing 10% heat-inactivated human serum, supplemented with streptomycin (100 Ag/ml),penicillin (100 units/ml), and 2-mercaptoethanol (50 1.M). After 5 days of coculture in a humidified CO2 incubator, the cellswere collected and their cytotoxic activity was determined by 51Cr-release assay.*Effector/target cell ratio.tSpleen cells were passed through a nylon-wool column and used as responder cells.tMRL/l mice were irradiated and reconstituted with bone marrow cells of BALB/c nu/nu mice.
MRL/l (3 mo) and BXSB (>9 mo) mice, cytotoxic T lym-phocytes were not generated, even when nylon-wool nonad-herent spleen cells were used as responder cells. It should benoted that T-cell dysfunction in autoimmune-prone mice isattributable not only to the involution of the thymus but alsoto abnormal stem cells, because normal stem cells ofBALB/c nu/nu mice can differentiate into functional T cellseven after the thymus of the autoimmune-prone recipient hasalready undergone considerable involution in BXSB orMRL/l strains.Thus, normal T cells, B cells, and macrophages have de-
veloped 3 mo after bone marrow transplantation of BALB/cnu/nu marrow cells into lethally irradiated MRL/l or BXSBmice. Based on these findings, we can conclude that the etio-pathogenesis of autoimmune diseases lies in a primary defectof stem cells and that abnormal stem cells induce prematurethymic involution and T-cell dysfunction. The abnormalstem cells also differentiate into abnormal macrophages andabnormal B cells; the latter then produce autoantibodieswith the help of abnormal macrophages and T cells. Thus, awide range of abnormalities consequent to these abnormali-ties in various compartments of the immune system appearsto induce the immunological imbalance that results in auto-immune diseases in mice of these autoimmune-prone strains.
In the present study, we used bone marrow cells of youngBALB/c nu/nu mice (<2 mo) because we have found that asmall number of functional T cells can be detected in spleenof nu/nu mice (12) and that these functional T cells increasewith age (25). The results described here suggest that alloge-neic bone marrow transplantation can be used without seri-ous reactions, even when MHC barriers are crossed, provid-ed that the T cells normally present in bone marrow are com-pletely removed. Candidates for treatments for autoimmunediseases include total lymphoid irradiation (26) and employ-ment of immunosuppressive agents such as cyclosporin A(27), azathioprine (28), cyclophosphamide (29), and/or ster-oid hormones. These approaches, however, exert suppres-sive effects on T-cell and other immunological functions.With the recent progress of therapy, most patients with auto-immune diseases do not die of renal failure but of opportu-nistic infection attributable to immunosuppression. Since weprovided evidence that T-cell functions increase after bonemarrow transplantation in mice, it seems likely with furtherimprovement in bone marrow transplantation technology,which will lead to reducing the mortality rate of the basicprocedure and selecting appropriate resistance genes forhaplo-identical donors, bone marrow transplantation ulti-mately might become a viable strategy for the treatment ofsevere life-threatening autoimmune diseases in humans.
We thank Dr. T. Tanisaka for performing irradiation studies. Wealso thank Mr. K. Tsuchida, Mr. T. Obata, Ms. K. Kitamura, Ms.M. Asano, and Ms. Myint Myint Than for their expert technical as-sistance and Ms. S. Kurimoto and Ms. K. Shimizu for their help inthe preparation of the manuscript. This work was supported in partby the following grants: a grant from the Japanese Ministry of Wel-fare and Health, the Research Grant for 1984 from the Dr. ShimizuFoundation for the Promotion of Immunology, the National Marchof Dimes-Birth Defects Foundation Grant 1-789, and grants AG-03592 and AI-19495 from the U.S. National Institutes of Health.
1. Andrews, B. S., Eisenberg, R. A., Theofilopoulos, A. N.,Izui, S., Wilson, C. B., McConahey, P. J., Murphy, E. D.,Roths, J. B. & Dixon, F. J. (1978) J. Exp. Med. 148, 1198-1215.
2. Krakauer, R. S., Waldman, T. A. & Strober, W. (1976)J. Exp.Med. 144, 662-673.
3. Cantor, H., McVay-Boudreau, L., Hugenberger, J., Naidorf,K., Shen, F. W. & Gershon, R. K. (1978) J. Exp. Med. 147,1116-1125.
4. DeHeer, D. & Edinton, T. S. (1978) J. Immunol. 118, 1858-1863.
5. McCombs, C., Hom, J., Talal, N. & Mischell, R. I. (1975) J.Immunol. 115, 1695-1699.
6. Warner, N. L. & Moore, M. A. S. (1971) J. Exp. Med. 134,313-334.
7. Datta, S. K., Manny, N., Andrzejewski, C., Andre-Schwartz,J. & Schwartz, R. S. (1978) J. Exp. Med. 147, 854-871.
8. Kincade, P. W., Lee, G., Fernandes, G., Moore, M. A. S.,Williams, N. & Good, R. A. (1979) Proc. Nad. Acad. Sci.USA 76, 3464-3468.
9. Dixon, F. J. (1982) Arthritis Rheum. 25, 721-725.10. Steinberg, A. D., Raveche, E. S., Laskin, C. A., Miller,
M. L. & Steinberg, R. T. (1982) Arthritis Rheum. 25, 734-743.11. Jyonouchi, H., Kincade, P. W., Good, R. A. & Fernandes, G.
(1981) J. Immunol. 127, 1232-1235.12. Ikehara, S., Pahwa, R. N., Fernandes, G., Hansen, C. T. &
Good, R. A. (1984) Proc. Nati. Acad. Sci. USA 81, 886-888.13. Sekita, K., Doi, T., Muso, E., Yoshida, H., Kanatsu, K. &
Hamashima, Y. (1984) Clin. Exp. Immunol. 55, 487-494.14. Koike, T., Nagasawa, R., Nagata, N. & Shirai, T. (1982) Im-
munol. Lett. 4, 93-97.15. Gershwin, M. E., Ikeda, R. M., Kruse, W. L., Wilson, F.,
Shifrine, M. & Spangler, W. (1978) J. Immunol. 120, 971-979.16. Theofilopoulos, A. N. & Dixon, F. J. (1981) Immunol. Rev.
55, 179-216.17. Ikehara, S., Tanaka, H., Nakamura, T., Furukawa, F., Inoue,
S., Sekita, K., Shimizu, J., Hamashima, Y. & Good, R. A.(1985) 0-Thymus, in press.
18. Izui, S., McConahey, P. J., Theofilopoulos, A. N. & Dixon,F. J. (1979) J. Exp. Med. 149, 1099-1116.
19. Theofilopoulos, A. N. (1982) Arthritis Rheum. 25, 726-733.20. Wofsy, D., Kerger, C. E. & Seaman, W. E. (1984) J. Exp.
Med. 159, 629-634.
Proc. NatL Acad Sci. USA 82 (1985)
Dow
nloa
ded
by g
uest
on
Oct
ober
1, 2
020
Medical Sciences: Ikehara et al.
21. Onod, K., Fernandes, G. & Good, R. A. (1980) J. Exp. Med.151, 115-132.
22. Krown, S. E., Coico, R., Scheid, M. P., Fernandes, G. &Good, R. A. (1981) Clin. Immunol. Immunopathol. 19, 268-283.
23. Kruisbeek, A. M., Sharrow, S. O., Mathieson, B. & Singer,A. (1981) J. Immunol. 127, 2168-2176.
24. Aizawa, S., Sado, T., Muto, M. & Kubo, E. (1981) J. Immu-nol. 127, 2426-2431.
Proc. NatL. Acad. Sci. USA 82 (1985) 2487
25. Slavin, S. (1979) Proc. Nad. Acad. Sci. USA 76, 5274-5276.26. Dominique, I. B., Noel, L. H., Bach, M. A., Dardenne, M. &
Bach, J. F. (1983) Clin. Exp. Immunol. 54, 359-365.27. Hahn, B. H., Kantor, 0. S. & Osterland, C. K. (1975) Ann.
Intern. Med. 83, 597-605.28. Shiraki, M., Fujiwara, M. & Tomura, S. (1984) Clin. Exp. Im-
munol. 55, 333-339.29. Fauci, A. S., Dale, D. C. & Balow, J. E. (1976) Ann. Intern.
Med. 84, 305-315.
Dow
nloa
ded
by g
uest
on
Oct
ober
1, 2
020
