Proc. Natl. Acad. Sci. USA Vol. 93, pp. 8558-8562, August 1996 Immunology

Thymus transplantation, a critical factor for correction of autoimmune disease in aging MRL/+ mice NAOKI HOSAKA*, MASATO NOSEt, MASAHISA KYOGOKUt, NORIKAzu NAGATA*, SHIGEO MIYASHIMA*, ROBERT A. GOOD§, AND SUSUMU IKEHARA*¶ *First Department of Pathology, Kansai Medical University, Moriguchi City, Osaka 570, Japan; tFirst Department of Pathology, Tohoku University, Sendai, Miyagi 980-77, Japan; *Otsuka Pharmaceutical Co. Ltd., Ohtsu, Shiga 520-01, Japan; and §Department of Pediatrics, All Children's Hospital, University of South Florida, St. Petersburg, FL 33701 Contributed by Robert A. Good, May 1, 1996

ABSTRACT MRL/MP-+/+ (MRL/+) mice develop tion of certain otherwise highly resistant autoimmune diseases pancreatitis and sialoadenitis after they reach 7 months of such as those that occur in MRL/lpr mice and the autoimmune age. Conventional transplantation has been polyarthritis of NZB/KN mice (9, 10). Old female MRL/+ found to be ineffective in the treatment of these forms of mice (>7 months) frequently develop apparent autoimmune apparent autoimmune disease. Old MRL/+ mice show a pancreatitis and sialoadenitis (11). In the present study, we dramatic thymic involution with age. Hematolymphoid recon- have successfully treated this form of late-onset autoimmune stitution is incomplete when fetal liver cells (as a source of diseases in mice using a combination of FT grafts, fetal liver hemopoietic stem cells) plus fetal bone (FB; which is used to cell (FHC) transplants plus transplantation of fragments of recruit stromal cells) are transplanted from immunologically fetal bones (FBs). We show herein that allogeneic FT grafts normal C57BL/6 donor mice to MRL/+ female recipients. given in conjunction with FB grafts and FHCs regularly result Embryonic from allogeneic C57BL/6 donors was in dramatic correction of the manifestations of the apparent therefore engrafted along with either bone marrow or fetal autoimmune pancreatitis and sialoadenitis in older MRL/+ hematopoietic cells (FHCs) plus fragments of adult or fetal mice. bone. More than seventy percent of old MRL/+ mice (>7 months) that had been given a fetal thymus (FT) transplant MATERIAL AND METHODS plus either bone marrow or FHCs and also bone fragments survived more than 100 days after treatment. The mice that Mice and Antibodies. Female MRL/+ mice were bred and received FHCs, FB, plus FI from allogeneic donors developed maintained in the animal facilities at the 1st Department of normal and B cell functions. Serum amylase levels Pathology, Tohoku University, Sendai, Japan, or obtained decreased in these mice whereas they increased in the mice from CLEA Japan, Osaka. Pregnant C57BL/6 mice were used that received FHCs and FB but not FT. The pancreatitis and as a source of donor cells, bone marrow, fetal liver, and fetal sialoadenitis already present at the time of transplantations cranial bones for transplantation. Fluorescein isothiocyanate were fully corrected according to histological analysis by (FITC)-coupled anti-H-2Kd or anti-H-2k antibodies were pur- transplants of allogeneic FHCs, FB and FT in the MRL/+ chased from the Meiji Institute of Health Science, Odawara, mice. These findings are taken as an experimental indication Kanagawa, Japan. that perhaps stem cell transplants along with FT grafts might Transplantation. Female MRL/+ mice (>7 months) were represent a useful strategy for treatment of autoimmune lethally (9.5 Gy) irradiated and divided into four groups. In diseases in aged humans. group 1 the mice were injected intravenously (i.v.) with 3-4 x 107 T cell-depleted BMCs from female C57BL/6 mice (2-3 During the past twenty years, remarkable advances have been months). These recipient mice were also engrafted with 2 to 3 made in bone marrow transplantation (BMT) as a form of stem fragments (2 x 2 x 3 mm) of bone from a femur from the same cell transplantation. Indeed, BMT has already become a donor mice. These bone grafts were placed under the renal powerful strategy for the treatment of leukemias, lymphomas, capsule of one kidney. In group 2, in addition to the transplants other cancers, aplastic anemias, congenital immunodeficien- of BMCs plus bone, the mice were transplanted with two cies, liposomal storage diseases, enzyme deficiencies, and thymic lobes from fetal (18-20 days of gestation) C57BL/6 genetically determined hematopoietic abnormalities. How- mice. The grafts were all placed under the same renal capsule. ever, the success of BMT in older patients is low because such In group 3, the mice were injected i.v. with 3-4 x 107 FHCs patients develop interstitial pneumonitis and acute or chronic as a source of hemopoietic stem cells and also engrafted with vs. host disease (GVHD) or show a relapse of the primary 2 to 3 fragments (2 x 2 x 3 mm) ofcranial bone (from the same disease for which the original transplant was performed (1-3). fetus) placed under the renal capsule (one side). In group 4, Thus far the reasons for these inadequacies of BMT in elderly two fetal thymic lobes from the same fetal donor were also patients have not been fully clarified. engrafted under the same renal capsule along with transplants We have previously found (4-8) that BMT can be used to of FB fragments and the i.v. injection of 3-4 x 107 FHCs. prevent or treat many spontaneous autoimmune diseases such Since it proved difficult to obtain large numbers of old as organ-specific autoimmunities like diabetes in NOD mice MRL/+ mice at any one time, several sets of experiments and systemic autoimmunities in (NZB x NZW)Fj, BXSB, and using relatively small groups of mice were performed; each (NZW x BXSB)Fj mice. We have recently shown that trans- group included at least 3 mice, and each experiment was plantation of both bone marrow cells (BMCs) and bones, repeated three or more times. Highly reproducible observa- which exhibit the capacity to recruit stromal cells from the same normal bone marrow donor, leads to complete preven- Abbreviations: FB, fetal bone; FT, fetal thymus; FHCs, fetal hema- topoietic cells (fetal liver); dGuo, deoxyguanosine; LPS, lipopolysac- charide; BMC, bone marrow cell; BMT, bone marrow transplantation; The publication costs of this article were defrayed in part by page charge GVDH, graft vs. host disease; Con A, concanavalin A; FITC, fluo- payment. This article must therefore be hereby marked "advertisement" in rescein isothiocyanate. accordance with 18 U.S.C. §1734 solely to indicate this fact. 'lTo whom reprint requests should be addressed. 8558 Downloaded by guest on October 3, 2021 Immunology: Hosaka et al. Proc. Natl. Acad. Sci. USA 93 (1996) 8559

100

0-%

4-01) L --- 50 L.

_ , *_ ...... ____I...... ,

V) 0 0 20 40 60 80 100 Survival days FIG. 1. Survival rate of mice in four groups. Lethally irradiated female MRL/+ mice (7 months) were-injected i.v. with 3-4 x lO7 T cell-depleted C57BL/6 BMCs and engrafted also with fragments of bone (group 1, ), or further engrafted with FT (group 2, ). Irradiated MRL/+ mice injected i.v. with 3-4 x 107 FHCs and engrafted with FB but no thymus graft (group 3, - - -), or irradiated female MRL/+ mice injected i.v. with 3-4 x 107 FHCs, FB grafts and also engrafted with FT (group 4, _).

tions were obtained and, thus, representative data are shown RESULTS in the figures. Assessment of Histopathology. Histopathological analyses Survival Rate. To treat autoimmune pancreatitis and si- aloadenitis in old female MRL/+ mice, we carried out BMT on were performed sections of the pancreas and salivary glands plus other tissue transplants using four methods. In group 1, after fixing and staining the tissues with hematoxylin and eosin old female MRL/+ mice (>7 months) were lethally irradiated using standard methods. and then injected i.v. with T cell-depleted BMCs. Mice of this Flow Cytometry. cells (1 x 106) were suspended in group were engrafted with bone fragments (femur) free of phosphate-buffered saline (PBS) containing 2% fetal calf BMCs from adult C57BL/6 mice (transplantation of adult serum (FCS) plus 0.05% sodium azide, then incubated on ice BMCs plus bone grafts) as described (9, 10). In group 2, in with appropriate antibodies for 30 min. All surface antigens addition to these transplants, the mice were engrafted with were stained by a direct immunofluorescence method using C57BL/6 FT (transplantation of adult T cell-depleted BMCs, FITC-coupled antibodies. The cells were analyzed using a adult bone plus FT). In group 3, the mice were injected i.v. with FACScan (Becton Dickinson). C57BL/6 fetal hematopoietic stem cells (FHCs) obtained Mitogen Responses. The proliferative responses of spleen from fetal liver at 18-20 days gestation and also engrafted with cells were evaluated as previously described (4). Briefly, the C57BL/6 cranial FB (transplantation of FHCs plus FB). In spleen cells (2 x 105) were cultured for 72 hr in 96-well group 4, the mice were treated as in group 3 but, in addition, flat-bottomed microculture plates with 50 ,ug/ml of lipopoly- engrafted with the C57BL/6 FT from the same embryo as the saccharide (LPS, Difco), 5.0 of concanavalin A (Con A, donor (transplantation of FHCs, FB, and FT). As shown in Fig. gg/ml the survival rate of the mice in group 1 at 100 was 10%. Sigma) in RPMI 1640 medium containing 10% FCS, 5 x 10-5 1, days In group was at 100 In M and 2 mM 2, survival =70% days. contrast, 2-mercaptoethanol glutamine. [3H]-thymidine approximately 20% of the mice in group 3 survived at 100 days, = x was to (0.5 m Ci; 1 Ci 3.7 1010 Bq) introduced each culture while the survival rate in group 4 was -80% after the several 16 hr prior to termination and proliferation then assessed. transplants. None of the mice in groups 2 or 4 showed clinical Measurement of Amylase. Mouse sera were diluted in PBS signs of GVHD or runting syndrome reminiscent of GVHD. (1:16) and the amylase levels then quantified using an ELISA We compared the mice in group 3 with those in group 4 to assay (available at Falco, Osaka). The data are expressed as analyze the benefits of thymus grafts, since the mice in group international units per liter. 1 died much earlier than those in the other groups. Statistical Analyses. Statistical analyses were performed H-2 Typing. To examine whether the hematopoietic cells of using Student's paired t test. the recipients were derived from donor cells, we carried out

A

L 0)~~~~~~~I 8 1. n.,

II' X

Fluorescence intensity FIG. 2. H-2 typing of spleen cells of mice in groups 3 (A) and 4 (B). The mice in groups 3 and 4 were sacrificed 2 months after transplantation. Cells were unstained ( ), stained with H-2b-FITC (-- -), and H-2k-FITC ( ) monoclonal antibodies. Downloaded by guest on October 3, 2021 8560 Immunology: Hosaka et al. Proc. Natl. Acad. Sci. USA 93 (1996)

6000

5000 _ Cz

4000 ;Ia i>

3000 .

i Before After Before After FIG. 3. Serum amylase levels in mice in groups 3 (A) and 4 (B) before and 2-3 months after transplantation. Statistical analyses performed using Student's paired t test revealed significant differences between the mice of these two groups. H-2 typing using spleen cells from the chimeric mice in groups Serum Amylase Levels. The amylase levels in the sera of the 3 and 4. As shown in Fig. 2A, more than 50% of the spleen cells mice in group 3 that were elevated initially did not fall after the of the mice in group 3 were H-2Kb-positive (donor-derived) fetal liver plus bone transplantations and actually increased. and approximately 30% were H-2Kk-positive (host-derived). These levels were 3906 ± 289 units/liter before the transplan- This finding suggests that the host hematopoietic cells had tation and 4339 ± 1115 units/liter after the transplantation recovered from the lethal irradiation and differentiated into (Fig. 3A). By contrast, the amylase levels of the mice in group mature cells within 2 months of BMT. By contrast, in the mice 4 showed significant reduction following the transplantation; of group 4, almost all of the spleen cells were derived from the 3792 ± 242 units/liter before transplantation and 3201 ± 488 donor hemopoietic cells. Only a few cells (3%) appeared to be units/liter following transplantation (P < 0.05) (Fig. 3B). derived from the cells of the recipient major histocompatibility Histology. Histopathological findings in sections of the complex (Fig. 2B). pancreas and salivary glands in nontreated female MRL/+

- - I 0 -. T31 _ T h1A-o FIG. 4. Representative histopathological findings of the pancreas and salivary glands in the nontreated female (12 months) MRL/+ mice (A and B) or mice (15 months) from group 4 (C and D). (X20.) Downloaded by guest on October 3, 2021 Immunology: Hosaka et al. Proc. Natl. Acad. Sci. USA 93 (1996) 8561

C57BL/6 (2 mo) C57BL/6 (14 mo) MRLU+ (2 mo) MRLU+ (13 mo) FHSCs + FB (1.5 mo later) FHSCs + FB + FT (4 mo later) FCA"" 10 20 30 Thymidine incorporation (cpm X10 ) FIG. 6. Mitogen responses in spleen cells of mice from groups 3 and 4. Spleen cells were cultured with medium alone (O), Con A (5 mg/ml) (m) or LPS (50 mg/ml) (u). state of the hematolymphoid cells (as shown in Fig. 2), this immunologic deficiency may be attributed either to failed reconstitution or to allogeneic cell-cell interactions. In con- trast, the mice in group 4 showed vigorous responses to both Con A and LPS, the levels observed being comparable to those of normal C57BL/6 mice.

DISCUSSION In the present study we performed allogeneic stem cell-based reconstitution following to determine which of the four methods might be the most advantageous approach to treating autoimmune disease in older MRL/+ mice. Almost all the mice in group 1 (transplantation of adult BMCs and bone) died within 2 wk of transplantation, while the mice in group 3 (transplantation of FHCs and FB), although surviving somewhat longer, did not achieve high survival rates over the 100 days after transplantation. By contrast, the mice in group 2 (transplantation of adult BMCs, bone fragments 4 of FHCs, FB, plus FTs) FIG. 5. Histology of bone marrow engrafted under the renal capsule. and FT) and group (transplantation The bone engrafted under the renal capsule of the mouse in group 3 survived much longer (Fig. 1). These findings indicate that (1) contains very few cells (A). Complete and vigorous reconstitution in mice given either adult BMCs plus bone fragments or FHCs the bone (right) and thymus (left) is noted in the mouse in group 4 (B) plus FB do not survive long, and that the transplantation of FT reconstituted with transplants of FHCs, FB, and FT. (X40.) is essential to rescue the old mice of this strain after stem cell transplantation. In analyses of the ability to reconstitute mice (12 months) are shown in Fig. 4 A and B. Impressive donor-type cells, we found that the spleen cells of the mice in infiltration of lymphoid cells was regularly apparent in the group 4 were almost all donor-derived, while a significant exocrine component of the pancreas of the untreated mice. number of host-type cells could be found in the of the Only hypertrophy of the islets without insulitis was noted in mice in group 3 (Fig. 2). Furthermore, the thymus showed a these control specimens (Fig. 4A). Focal lymphocyte infiltra- normal structure, and the transplanted bones strongly sup- tion was also present around the ducts of the salivary glands of in but not in those the ported hematopoiesis in the mice group 4, the untreated mice (Fig. 4B). The histological findings of in group 3 (Fig. 5). This finding suggests that FT transplan- mice in group 4,3 months following transplantation, are shown of donor- D. infiltration had disappeared tation facilitates the development and engraftment in Fig. 4 C and Lymphocyte type hematopoietic cells, probably by suppressing the differ- from both pancreas (C) and salivary glands (D) with irradia- that the tion plus transplantation of FHCs, bone fragments and thymus entiation of host-derived cells. It is therefore likely lobes. As shown in Fig. 5, the bone engrafted under the renal donor-derived T cells induced by FT grafts may play a crucial role in the of graft rejection that might otherwise capsule in the mice in group 3 contained only a few cells. These suppression findings reflected the incomplete reconstitution of hemato- be mediated by host-derived cells such as T cells, natural killer observed in mice that had not been given a thymus (Fig. cells, and macrophages. poiesis the 5A). By contrast, the mice from group 4 that received FHCs Concerning the treatment of autoimmune diseases, plus FB and FT showed a normal appearance in the hemato- chronic pancreatitis and sialoadenitis that characterizes the poietic tissue, reflecting the complete hematopoietic recon- MRL/+ mice appear to have been regularly cured in the mice stitution achieved (Fig. SB). in groups 2 (data not shown) and 4 (Figs. 3 and 4). Immunologic Functions. T cell and B cell functions were The spleen cells in the mice in group 4 showed a vigorous evaluated using responses to mitogens as indicators (Fig. 6). proliferative response to the mitogens (Con A and LPS) at a The spleen cells of old MRL/+ mice (13 months) showed a level comparable to that of nontreated adult C57BL/6 mice lower response to T cell mitogen (Con A) than young MRL/ + (Fig. 6). In humans, it has been reported that the major causes mice. By contrast, the response to the B cell mitogen (LPS) of of death after BMT in old recipients (>45 years old) are these MRL/+ mice did not change with age (Fig. 6). The interstitial pneumonitis, acute GVHD, and relapse of the spleen cells of mice from group 3 (FHCs and FB) that had primary disease for which hematopoietic transplantation was survived showed decreases in both Con A and LPS responses, carried out. Except for acute GVHD, these causes of death reflecting the incomplete reconstitution of lymphoid cells in appear to be due to incomplete reconstitution followed by these mice. Since the mice in group 3 showed a mixed chimeric immunodeficiency or graft rejection. Downloaded by guest on October 3, 2021 8562 Immunology: Hosaka et al. Proc. Natl. Acad. Sci. USA 93 (1996)

There have been numerous reports concerning thymus 3. Ringden, 0. & Nilsson, B. (1985) Transplantation 40, 39-44. grafts in humans (13-15). Most of these contain negative or 4. Ikehara, S., Good, R. A., Nakamura, T., Sekita, K., Inoue, S., unimpressive data that are probably attributable either to Maung Maung O., Muso, E., Ogawa, K. & Hamashima, Y. (1985) rejection of the engrafted thymus by radioresistant host cells Proc. Natl. Acad. Sci. USA 82, 2483-2487. or the development of GVHD by T cells contaminating the 5. Ikehara, S., Ohtsuki, H., Good, R. A., Asamoto, H., Nakamura, T., Sekita, K., Muso, E., Tochino, Y., Ida, T., Kuzuya, H., Imura, engrafted thymus (13, 14). However, we have recently exam- H. & Hamashima, Y. (1985) Proc. Natl. Acad. Sci. USA 82, ined the cytotoxic effects of irradiation and deoxyguanosine 7743-7747. (dGuo) on the thymus and found that FT is sensitive to both 6. Yasumizu, R., Sugiura, K., Iwai, H., Inaba, M., Makino, S., Ida, irradiation and dGuo (>1.35 mm) (16). In addition, we have T., Imura, H., Hamashima, Y., Good, R. A. & Ikehara, S. (1987) found that the transplantation of allogeneic FT without irra- Proc. Natl. Acad. Sci. USA 84, 6555-6557. diation or dGuo treatment leads to long-term T cell reconsti- 7. Oyaizu, N., Yasumizu, R., Miyama-Inaba, M., Nomura, S., tution in nu/nu mice, and that the nu/nu mice do not develop Yoshida, H., Miyawaki, S., Shibata, Y., Mitsuoka, S., Yasunaga, GVHD even when the allogeneic FT used has contained K., Morii, S., Good, R. A. & Ikehara, S. (1988) J. Exp. Med. 167, immature at the time of engraftment (16). Based 2017-2022. on these findings and those observed in the present report, we 8. Ikehara, S., Yasumizu, R., Inaba, M., Izui, S., Hayakawa, K., propose Sekita, K., Toki, J., Sugiura, K., Iwai, H., Nakamura, T., Muso, would that the transplantation of the FT (without E., Hamashima, Y. & Good, R. A. (1989) Proc. Natl. Acad. Sci. using irradiation or dGuo treatment) together with FHCs plus USA 86, 3306-3310. FB might be appropriate as a strategy to treat patients older 9. Ishida, T., Inaba, M., Hisha, H., Sugiura, K., Adachi, Y., Nagata, than 45 yr who suffer from a variety of diseases, including N., Ogawa, R., Good, R. A. & Ikehara, S. (1994)J. Immunol. 152, life-threatening autoimmune diseases. 3119-3127. 10. Nakagawa, T., Nagata, N., Hosaka, N., Ogawa, R., Nakamura, K. The authors thank Ms. Y. Shinno and Ms. Y. Matsui for their & Ikehara, S. (1993) Arthritis Rheum. 36, 263-268. technical assistance, Ms. K. Ando for her help in the preparation of the 11. Kanno, H., Nose, M., Itoh, J., Taniguchi, Y. & Kyogoku, M. manuscript, and Ms. Tazim Verjee for both manuscript preparation (1992) Clin. Exp. Immunol. 89, 68-73. and editing. This work was supported in part by a grant for Experi- 12. Phan, D. T., Phuong, D. T., An, V. T., Khue, P., Petranyi, C. G. mental Models for Intractable Diseases from the Ministry of Health & Hollan, S. R. (1987) Thymus 9, 123-125 (lett.). and Welfare of Japan, and grants-in-aid for scientific research 13. Atkinson, K., Storb, R., Ochs, H. D., Goehle, S., Sullivan, K. M., 02152117, 02670162, and 03454177 from the Japanese Ministry of Witherspoon, R. P., Lum, L. G., Tsoi, M. S., Sanders, J. E., Parr, Education, Science, and Culture. M., Stewart, P. & Thomas, E. D. (1982) Transplantation 33, 168-173. 1. Klingemann, H. G., Storb, R., Fefer, A., Deeg, H. J., Appelbaum, 14. Pahwa, S., Pahwa, R., Incefy, G., Reece, E., Smithwick, E., F. R., Buckner, C. D., Cheever, M. A., Greenberg, P. D., Stewart, O'Reilly, R. & Good, R. A. (1979) Clin. Immunol. Immuno- P. S., Sullivan, K. M., Witherspoon, R. P. & Thomas, E. D. (1986) pathol. 14, 107-120. Blood 67, 770-776. 15. Mayumi, M., Kimata, H., Suehiro, Y., Hosoi, S., Ito, S., Kuge, Y., 2. Weiner, R. S., Bortin, M. M., Gale, R. P., Gluckman, E., Kay, Shinomiya, K. & Mikawa, H. (1989) Eur. J. Pediatr. 148,518-522. H. E. M., Kolb, H. J., Hartz, A. J. & Rimm, A. A. (1986) Ann. 16. Kumamoto, T., Inaba, M., Toki, J., Adachi, Y., Imamura, H. & Intern. Med. 104, 168-175. Ikehara, S. (1995) Immunobiology 192, 365-381 Downloaded by guest on October 3, 2021