CD4+ −Depleted Lymphocyte Infusion Impairs Neither the Recovery of Recipient nor the Development of Transplanted Thymus This information is current as of September 26, 2021. Ming Shi, Ming Li, Yunze Cui, Lin Liu, Yasushi Adachi and Susumu Ikehara J Immunol 2013; 190:2976-2983; Prepublished online 4 February 2013; doi: 10.4049/jimmunol.1201605 Downloaded from http://www.jimmunol.org/content/190/6/2976

Supplementary http://www.jimmunol.org/content/suppl/2013/02/04/jimmunol.120160 http://www.jimmunol.org/ Material 5.DC1 References This article cites 36 articles, 10 of which you can access for free at: http://www.jimmunol.org/content/190/6/2976.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 26, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Author Choice Freely available online through The Journal of Immunology Author Choice option Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2013 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CD4+ T Cell–Depleted Lymphocyte Infusion Impairs Neither the Recovery of Recipient Thymus nor the Development of Transplanted Thymus

Ming Shi,* Ming Li,* Yunze Cui,*,† Lin Liu,* Yasushi Adachi,*,‡ and Susumu Ikehara*

Thymus transplantation, in conjunction with transplantation (BMT), has been attracting attention for the treatment of various diseases. Recently, donor lymphocyte infusion (DLI) has been used as a helpful tool for establishing donor chimerism and preventing a relapse of leukemia/lymphoma. However, the effects of DLI on transplanted and recipient thymuses have not been explored. We therefore performed DLI in the intrabone marrow–BMT + thymus transplantation setting. We have found that DLI leads to derangements in both recipient thymuses and transplanted thymuses; by 2 wk after BMT, we saw a decrease in total cell number, a lower percentage of CD4+CD8+ cells, and the obliteration of the thymic corticomedullary junction. Four weeks later, the Downloaded from thymic impairment became more serious. However, when we depleted the CD4+ T cells (CD42-DLI), the recipient thymic recovery and transplanted thymic development were significantly restored by the treatment. In addition, there were much greater levels of TNF-a and Fas ligand, and a lower percentage of regulatory T cells in the DLI group than in the CD42-DLI group. These findings indicate that inflammation induced by DLI, especially by CD4+ T cells, plays a crucial role in the thymic impairment. The Journal of Immunology, 2013, 190: 2976–2983. http://www.jimmunol.org/ llogeneic bone marrow transplantation (allo-BMT) is chronic GvHD (13). Thymus transplantation (TT), an attractive a potentially curative therapy for certain diseases of the method for improving T cell functions, has been applied clinically A hematopoietic system, immunodeficiencies, autoimmune for patients with DiGeorge syndrome or HIV infection, which diseases, solid malignant tumors, and so on (1–6). We have developed elicits the hypoplasia of the thymus (14). However, in mice, al- a new and powerful bone marrow transplantation (BMT) method: though T cell functions were restored or enhanced by TT, no con- intrabone marrow–BMT (IBM-BMT) (7), in which donor bone comitant GvHD was observed after TT in conjunction with allo- marrow cells (BMCs) are directly injected into the recipient’s bone BMT (15). Therefore, TT can be used to treat autoimmune diseases marrow cavity. Therefore, a much greater number of donor hemato- in chimeric-resistant MRL/lpr mice and type 2 diabetes mellitus,

poietic stem cells and mesenchymal stromal cells (including mesen- and to suppress tumor growth (16–18). by guest on September 26, 2021 chymal stem cells) can be inoculated into the recipient bone marrow Donor lymphocyte infusion (DLI) is often used after allo-BMT by IBM-BMT than by conventional i.v. BMT. This results in the rapid to prevent disease relapse in the setting of T cell–depleted BMT or reconstitution of donor hematopoietic cells and permits a reduction nonmyeloablative conditioning regimens. It is also a combined in the doses of irradiation used as a conditioning regimen (8–10). method to convert from mixed chimerism to full donor chimerism The thymus is an organ for inducing T cells and maintaining (19, 20). However, DLI-induced GvHD is always associated with homeostasis. However, thymic functions are impaired by the con- an increase in therapy-related morbidity because of its uncon- ditioning regimen and the acute -versus-host disease (GvHD) trollable and fatal characteristics (21). It has been reported that that occurs after allo-BMT, resulting in deficient cell immunity (11, many factors are involved in the damage to the recipient thymus 12). In addition, there is a strong association between posttrans- after DLI (22, 23), whereas the effects of DLI on the transplanted plant autoimmune disease and the thymic dysfunction caused by thymus have hitherto remained unexplored. In this study, we investigate the influence of DLI on both re-

*Department of Stem Cell Disorders, Kansai Medical University, 570-8506 Osaka, cipient and transplanted thymuses in the IBM-BMT + TT setting. Japan; †JIMRO Co., Ltd., 370-0021 Gunma, Japan; and ‡Division of Clinical Pathol- Because we have found that TT using newborn thymus is most ogy, Toyooka Hospital, 668-8501 Hyogo, Japan effective in tumor suppression (18), we used newborn thymus in this Received for publication June 13, 2012. Accepted for publication January 3, 2013. study. We show in this article that CD4+ T cell–depleted lympho- 2 This work was supported by the Otsuka Pharmaceutical Company, Ltd., and by grants cyte infusion (CD4 -DLI) impairs neither the recovery of recipient from the Research on Allergic Disease and Immunology Committee from the Health thymus nor the development of transplanted thymus. and Labour Sciences Research of the Ministry of Health, Labour and Welfare. Address correspondence and reprint requests to Prof. Susumu Ikehara, Department of Stem Cell Disorders, Kansai Medical University, 10-15 Fumizono-cho, Moriguchi Materials and Methods City, Osaka 570-8506, Japan. E-mail address: [email protected] Mice The online version of this article contains supplemental material. C57BL/6 (B6), enhanced GFP (eGFP) transgenic (tg) B6, and BALB/c mice Abbreviations used in this article: allo-BMT, allogeneic bone marrow transplantation; were purchased from Shimizu Laboratory Supplies (Shizuoka, Japan). B6, C57BL/6; BMC, bone marrow cell; BMT, bone marrow transplantation; DLI, Eight- to 12-wk-old male mice were used for BMT and DLI. For TT, donor lymphocyte infusion; eGFP, enhanced GFP; FasL, Fas ligand; GvHD, graft- 1 d after birth, B6 mice were sacrificed to obtain newborn thymuses. All the versus-host disease; IBM-BMT, intrabone marrow–BMT; tg, transgenic; Treg, regu- latory T cell; TT, thymus transplantation; WSP, whole . mice were maintained in a specific pathogen-free room. This article is distributed under The American Association of Immunologists, Inc., Experimental protocol Reuse Terms and Conditions for Author Choice articles. As shown in Fig. 1, BALB/c mice were lethally irradiated with 7 Gy using Copyright Ó 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00 the Gammacell 40 Exactor (MDS Nordion, Kanata, ON, Canada) with two www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201605 The Journal of Immunology 2977

[137Cs] sources, and the next day, these mice received IBM-BMT from B6 some mice were injected i.p. with 1 mg anti–TNF-a Ab and 2 mg anti-FasL mice (group I). Some mice additionally received TT from B6 mice (group Ab (WSP + Abs group) on days 0, 4, 8, and 12 as described previously (26). II). On the same day, some mice also received whole spleen (WSP-), CD42-, or CD82-DLI from B6 mice: WSP-DLI (group III), CD42-DLI Statistical analysis (group IV), and CD82-DLI (group V). The treated mice were sacrificed The results are represented as means 6 SD. The Student t test was used to 5 d, 2 wk, or 4 wk after the treatments. determine any statistical significance. A p value ,0.05 was considered to Reagents and flow cytometric analysis be a significant difference. The Abs used in this study were as follows: purified rat anti-mouse CD4 and CD8 Ab (eBioscience, San Diego, CA); FITC-conjugated anti-mouse Results CD4 and H-2Kb Ab; PE-conjugated anti-mouse H-2Kd, CD4, CD8, and In the IBM+TT setting, WSP-DLI induces serious GvHD but B220 Ab; and PerCP-Cy5.5–conjugated anti-mouse CD45 Ab (BD Phar- CD42 -DLI does not mingen, San Diego, CA). For the thymus and peripheral blood analysis, leukocytes were first gated by CD45+ cells, which were estimated as nu- We carried out IBM-BMT, TT, and DLI from B6 to BALB/c mice. clear cells. To analyze the percentage of regulatory T cells (Tregs) in the The protocol is described in detail in Fig. 1 and in Materials and spleen, we performed intracytoplasmic FoxP3 staining using an eBio- Methods. science FITC–anti-mouse/rat FoxP3 staining set in accordance with the manufacturer’s instructions. Samples were analyzed using an FACSCalibur The TT group (group II) showed similar results to group I (Fig. flow cytometer (BD Biosciences). 2), although TT provides an abundance of (immature T cells), whereas DLI contains mature T cells and attacks the IBM-BMT and TT target organs after priming. Serious GvHD was induced not only One day after irradiation, the BMCs were prepared by flushing them from

according to the clinical scores (Fig. 2A), but also according to the Downloaded from the medullary cavities of the femurs and tibias of B6 mice with PBS. The 3 7 histopathological evaluation (Fig. 2B) in the WSP-DLI group BMCs (1 10 ) were then injected directly into the tibial cavity of the + recipient mice via the IBM route, as previously described (24). One whole (group III). However, when CD4 T cells were depleted, the severity + newborn thymus from a B6 mouse was transplanted under the renal cap- of the GvHD decreased dramatically (group IV), whereas CD8 sule of each recipient of IBM-BMT. T cell–depleted DLI (CD82-DLI: group V) induced GvHD, as seen Cell preparation and DLI in the WSP-DLI group (group III). Therefore, we omitted group V from the subsequent experiments because in vivo experiments Whole splenocytes from donor B6 mice were injected via the tail vein in the 2 http://www.jimmunol.org/ WSP-DLI group. CD4+ or CD8+ T cells were depleted from the spleno- require a large number of mice. Thus, the mice with CD4 -DLI cytes using purified CD4 or CD8 Ab with Dynabeads anti-rat IgG (Invi- exhibited only a slight and transient loss of body weight, but no trogen Dynal AS, Oslo, Norway) for the CD42- or the CD82-DLI group, serious GvHD (Fig. 2). respectively. A total of 1 3 107 splenocytes were injected in the WSP-, CD42-, and CD82-DLI groups. The recovery of recipient thymus is impaired by WSP-DLI but not CD42-DLI Assessment of GvHD The cytoreductive conditioning regimens in the context of BMT, Survival was monitored daily. The severity of GvHD was assessed as previously described (25). In brief, the recipients were scored every 4 or especially irradiation, result in severe recipient thymic atrophy and

5 d for six clinical parameters: weight loss, posture, activity, fur texture, defects in the (13). Moreover, after BMT, the by guest on September 26, 2021 skin integrity, and diarrhea. A severity scale of 0–2 was used for each transplanted bone marrow–derived thymic progenitors migrate to parameter, with a maximum score of 12. Histopathological analysis was the recipient’s thymus where they support thymopoiesis. We ex- performed by scoring the changes in the skin, intestine, and liver speci- mens. A severity scale from 0–4 was used, with a maximum score of 12. amined the effects of DLI on the recovery of the recipient thymus. The recipient mice were sacrificed and analyzed 5 d, 2 wk, and 4 ELISA wk after BMT. Five days after BMT, the recipient thymuses in all Peripheral blood was collected from the recipient mice by EDTA tube (BD, groups showed severe atrophy (Fig. 3A) and the total cell number Franklin Lakes, NJ) and centrifuged to get the mouse plasma within 30 min. decreased. The cell numbers in the WSP- and CD42-DLI groups The plasma was stored at 220˚C until assay. TNF-a and Fas ligand (FasL) were much lower than in the TT group (Fig. 3B). The size and were measured using the commercial kit (R&D Systems, Minneapolis, MN). Samples and standards were run in duplicate. weight of the recipient thymus in the TT group gradually in- creased (Fig. 3A, 3C, 3D). Histologically, the recipient thymus Blockade of inflammatory cytokine already displayed well-defined cortical and medullary areas 2 wk + + For cytokine blockade, purified rat anti-mouse TNF-a Ab and FasL Ab were after BMT (Fig. 3E). In addition, the percentage of CD4 CD8 purchased from BioXCell and BioLegend, respectively. After WSP-DLI, cells had also increased to the normal level (Fig. 3F).

FIGURE 1. Experimental protocol. BALB/c mice were lethally irradiated (7 Gy from [137Cs]). The next day, all the mice received IBM-BMT from B6 mice (group I). Some mice additionally received newborn TT from B6 mice (group II). On the same day, some mice also received WSP-, CD42-, or CD82-DLI from B6 mice: WSP-DLI (group III), CD42-DLI (group IV), and CD82-DLI (group V). The treated mice were sacrificed 5 d, 2 wk, or 4 wk after the treatments. 2978 THYMUS IMPAIRMENT BY CD4+ T CELLS

FIGURE 2. Clinical and histopathological scores of different groups. Irradiated BALB/c mice received IBM-BMT alone (BMT) or with newborn TT. Whole, CD4+, or CD8+ T cell–depleted splenocytes were injected in the WSP-, CD42- and CD82-DLI groups, respectively. Donor BMCs, thymuses, and splenocytes were all from B6 mice. After BMT, the body weight of the recipients was recorded and clinical signs of GvHD were assessed every 4 or 5 d (A). Autopsies were performed on mice that had been sacrificed 4 wk after BMT. Tissues from GvHD target organs (liver, skin, and intestine) were prepared for histopathological scoring (B). n = 4–8/group; experiments were performed three times. **p , 0.01.

However, when the whole B6 splenocytes were injected, the earlier time point. It has been reported that TT can also signifi- process of recovery was disturbed: 1) the recipient thymus re- cantly increase the percentage of CD4+ T cells (but not CD8+ mained atrophic (Fig. 3A, 3C, 3D); 2) it was structurally damaged T cells) compared with BMT alone (27). Because no serious Downloaded from and, histologically, lost the corticomedullary junction (Fig. 3E); GvHD or thymus impairment was induced by TT, we hypothe- and 3) the percentage of CD4+CD8+ cells decreased seriously sized that the derivation of CD4+ T cells could be the key point. (Fig. 3F, 3G). We used splenocytes of eGFP tg B6 mice to distinguish the der- When CD4+ T cells were depleted, the effects of DLI on the ivation of the T cells and confirm the role of the CD4+ T cells. In impairment of the recipient thymus appeared to be ameliorated at this model, BMCs and newborn thymus were from B6 mice,

2 wk after the BMT (Fig. 3C, 3E, 3F). Four weeks after the BMT, whereas splenocytes from eGFP tg mice were injected for DLI. http://www.jimmunol.org/ there was no significant difference in the structure or percentage of The percentages of non-DLI (donor BMCs or TT) and DLI- CD4+CD8+ cells between the TT and CD42-DLI groups (Fig. 3E, derived CD4+ and CD8+ T cells were analyzed 5 d after the 3G). These findings suggested that DLI, especially with CD4+ BMT. The results showed that the percentage of DLI-derived allo- T cells, impaired the recovery of the recipient thymus. reactive CD4+ T cells was much higher in the WSP-DLI group than in the CD42-DLI group, suggesting that the DLI-derived The development of transplanted thymus is also impaired by CD4+ T cells induced the impairment of both the recipient and WSP-DLI but not CD42-DLI transplanted thymus (Fig. 5A). Although the percentage of CD8+ The transplanted thymus was also examined at the time points T cells in both the WSP- and CD42-DLI groups was higher than described earlier. In the TT group (group II), macroscopic findings in the TT group, there was no significant difference between the by guest on September 26, 2021 indicated that the transplanted thymus developed well in all the WSP- and CD42-DLI groups, indicating that the CD8+ T cells recipient mice (Fig. 4A). Two weeks after the BMT, the trans- were likely not responsible for the thymus impairment (Fig. 5B). planted thymus showed normal architecture and percentage of Moreover, there were no significant differences in the percentages CD4+CD8+ cells (Fig. 4C, 4D). The thymus had also grown large of non–DLI-derived CD4+ and CD8+ T cells between the WSP- by 4 wk after BMT (Fig. 4A, 4C). In the WSP-DLI group, the total and CD42-DLI groups (Fig. 5A, 5B). These data showed that cell number of the transplanted thymus was lower than that of the mature CD4+ T cells from DLI played an important role in the TT group on day 5 after BMT, but there was no significant dif- impairment of the recovery of the recipient thymus and the de- 2 ference between the TT and CD4 -DLI groups (Fig. 4B). In the velopment of the transplanted thymus. In addition, lymphopenia, WSP-DLI group, only 37.5% of transplanted thymuses could be especially B lymphopenia, was seen in the WSP-DLI group, whereas observed at 2 wk after the BMT, and those thymuses were smaller, normal reconstitution was observed in the TT and CD42-DLI had lost the thymic corticomedullary junction, and had a de- groups at 4 wk after BMT (Table I), proving that donor CD4+ (not creased percentage of CD4+CD8+ cells (Fig. 4A, 4C, 4D). The CD8+) T cells were responsible for the induction of hematopoietic impairment in the transplanted thymus had become even more niche injury (28). The recipients in all groups showed full donor serious by the later time point, at which time only 16.7% of the chimerism (.98%, data not shown). transplanted thymuses could be observed (Fig. 4A, 4C, 4E). In + 2 Mechanism for CD4 T cell–induced thymic impairment contrast, 2 wk after BMT, the transplanted thymus in the CD4 -DLI group was much larger than that in the WSP-DLI group. In addi- It has been shown that alloreactive CD4+ T cells induce epithelial tion, both the thymic structure and the percentage of CD4+CD8+ and organ damage in recipients, otherwise known as acute GvHD; cells were similar to those in the TT group. Four weeks after BMT, many kinds of soluble inflammatory agents, such as TNF-a, IFN- there were no significant differences in any of the thymic param- g, IL-1, and NO, are involved (29). The target organs in acute eters between the TT and CD42-DLI groups (Fig. 4A, 4C–E). GvHD include the skin, liver, intestine, and thymus. Teshima and These results indicated that the WSP-DLI had a remarkable im- colleagues (30) have shown that the damage does not require pact on the development of the transplanted thymus, and that the alloantigen expression on the target epithelium, and that the neu- CD4+ T cells played a crucial role in the impairment. tralization of TNF-a prevents the damage. In addition, FasL-

+ dependent donor T cell–mediated cytotoxicity has been implicated DLI-derived CD4 T cells induce thymic impairment in the development of the thymus and bone marrow GvHD (22, 28). Because both the recovery of the recipient’s thymus and the de- Therefore, we hypothesized that the inflammatory cytokines of velopment of the transplanted thymus had already been disturbed TNF-a and FasL should induce the impairment of both recipient in the WSP-DLI group by 2 wk after BMT, we investigated the and transplanted thymuses. We thus examined the plasma levels of origin of the population that played a role in the impairment at an TNF-a and FasL 5 d after BMT, because we know that, at that The Journal of Immunology 2979 Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021

FIGURE 3. WSP-DLI, but not CD42-DLI, induces recipient thymic impairment. Recipient mice were sacrificed for analysis 5 d, 2 wk, or 4 wk after BMT. Thymuses from normal B6 mice and B6 mice 1 d after irradiation (7 Gy, no BMT) were used for control. (A) Representative macroscopic findings. Yellow circles indicate location of recipient thymus. (B) Total cell number of recipient thymus on day 5. (C–G) Weight, representative histological findings, and CD4+CD8+ cell percentage of recipient thymus are shown for later time points [(C, F): 2 wk; (D, G): 4 wk]. (E) H&E staining (original magnification 3100). n = 4–8/group; experiments were performed three times. **p , 0.01. time point, the processes of recipient thymic recovery and trans- impairment, according to observations made 4 wk after BMT (data planted thymic development were disturbed by WSP-DLI. The not shown). These results proved that the inflammatory cytokines levels of TNF-a and FasL were, indeed, significantly higher in the of TNF-a and FasL mediated the impairment of both recipient and WSP-DLI group than in the TT and CD42-DLI groups on day 5 transplanted thymuses. after BMT (Fig. 6A). To determine whether the production of We also compared the percentage of Tregs in the recipient inflammatory cytokines was causally related to the thymic im- spleen, which reflects the suppressive capability of inflammation pairment, we blocked TNF-a and FasL after WSP-DLI by Abs as 4 wk after BMT at the time lymphopenia occurs. The percentage of described previously (26). The mice injected with Abs were sac- Tregs in the WSP-DLI group was much lower than in the TT and rificed 2 or 4 wk after BMT. Both recipient and transplanted CD42-DLI groups (Fig. 6B). These data suggested that inflam- thymuses in the WSP + Abs group showed similar results to those mation induced by DLI-derived CD4+ T cells resulted in the im- in the CD42-DLI group 2 wk after BMT, including macroscopic pairment of recipient thymic recovery and transplanted thymic (Supplemental Fig. 1A) and histological (Supplemental Fig. 1B) development. observations, weight of recipient thymus (Supplemental Fig. 1C), and percentages of CD4+CD8+ cells in the recipient (Supple- Discussion mental Fig. 1D) and transplanted thymuses (Supplemental Fig. In this study, we examined the effects of CD42-DLI on both re- 1E). Blockade of TNF-a and FasL also prevented thymic cipient and transplanted thymuses after allo-IBM-BMT + TT. In 2980 THYMUS IMPAIRMENT BY CD4+ T CELLS Downloaded from http://www.jimmunol.org/

FIGURE 4. The development of transplanted thymus is impaired by WSP-DLI, but not CD42-DLI. Analysis of transplanted thymus was also performed by guest on September 26, 2021 at the same time as that in Fig. 3. (A) Representative macroscopic findings. Yellow circles indicate location of transplanted thymus. (B) Total cell number on day 5. (C) Representative histological findings. H&E staining (original magnification 3100). Plain arrows indicate cortex; dotted arrows indicate medulla. (D and E) Percentage of CD4+CD8+ cells is shown 2 or 4 wk after BMT. Each small square (n) represents the percentage for each recipient mouse in the left panels. Values of small squares on the x-axis are 0 and mean that no transplanted thymus could be observed or obtained for analysis. The right panels show the percentage by mean 6 SD. *p , 0.05, **p , 0.01. the BMT + TT setting, DLI induced serious damage, as evidenced paired, as was the development of the transplanted thymus. DLI in the classical GvHD target organs such as the skin, liver, and led to decreased total cell number, lower percentages of CD4+ intestine. In addition, recovery of the recipient thymus was im- CD8+ cells, and loss of the thymic corticomedullary junction in

FIGURE 5. Percentages of DLI- and non–DLI- derived CD4+ and CD8+ T cells 5 d after BMT. BMCs and newborn thymus from B6 mice were used for BMT and TT, whereas splenocytes from eGFP tg mice were injected for DLI. Peripheral blood was analyzed 5 d after BMT. The percentages of DLI- (eGFP+) and non–DLI-derived (eGFP2) CD4+ (A) and CD8+ (B) T cells in the leukocytes are shown. n = 3–4/group; experiments were performed three times. **p , 0.01. The Journal of Immunology 2981

Table I. Reconstitution of donor-derived hematopoietic cells 4 wk after derived hematopoietic stem cells, and help to suppress GvHD. BMT As shown in Fig. 2, GvHD was induced after WSP-DLI but was preventable with CD42-DLI. The process of recipient thymic re- Percentage of Donor Cells covery was disturbed, and the recipient thymus showed decreased + + Groups CD4 CD8 B220 total cell number, a lower percentage of CD4 CD8 cells, and obliteration of the thymic corticomedullary junction at 2 or 4 wk Group II (TT) 11.4 6 2.7 7.7 6 0.9 44.7 6 1.0 Group III (WSP) 2.4 6 1.1 4.7 6 1.4 2.0 6 1.2 after BMT (Fig. 3C–G). Because GvHD is considered to be sys- Group IV (CD42) 10.7 6 2.4** 7.0 6 1.3* 45.5 6 4.0** temic, the recipient thymus can also be considered to be a target for GvHD. Lower doses of DLI resulted in thymic damage without Four weeks after BMT, the reconstitution was analyzed using the peripheral blood by flow cytometer. Full donor chimerism was observed in all the recipient mice any obvious clinical signs of GvHD, suggesting that the thymus (.98%, checked by H-2Kb and H-2Kd staining). The percentages of H-2Kb+CD4+, is probably exquisitely sensitive to GvHD (22, 31). H-2Kb+CD8+, and H-2Kb+B220+ cells in the peripheral blood are shown. n =4–6/ group, experiments were performed three times. Effector mechanisms of acute GvHD are thought to include *p , 0.05, **p , 0.01, CD42-DLI group versus WSP-DLI group. There were no both CTL-mediated and cytokine-mediated cytotoxicity. In our significant differences in the percentages of donor-derived CD4, CD8, and B220 cells 2 model, recipient thymic impairment was significantly ameliorated between the TT and CD4 -DLI groups. 2 in the CD4 -DLI group, indicating that mature CD4+ T cells play a central role in the impaired recovery of the recipient thymus. We both the recipient and transplanted thymuses. However, the thymic also excluded the effects of TT-derived CD4+ T cells and DLI- impairment was significantly ameliorated after CD4+ T cell de- derived CD8+ T cells by using splenocytes from eGFP tg B6 mice pletion, indicating that CD4+ T cells play a crucial role in the (Fig. 5). Our findings of the relative importance of allo-reactive Downloaded from impairment. CD4+ T cells are consistent with recent reports showing that CD4+ Complete reconstitution of the recipient’s immune system is T cells are much more powerful (.100 times) than CD8+ T cells imperative for there to be a favorable outcome from allo-BMT. in destroying the hematopoietic niche (28), although it has been The recovery of peripheral T cells occurs via both thymus- reported that CD8+ T cells play a key role in mediating GvHD (33, independent and thymus-dependent pathways (31), and it seems 34). The mature T cells from DLI are activated and proliferate in

likely that mature T cells injected by DLI expand rapidly in re- the secondary lymphoid tissues under the stimulation of APCs. http://www.jimmunol.org/ sponse to recipient APCs and provide protective immunity after The activated CD4+ T cell–dependent GvHD damage is mediated allo-BMT via the thymus-independent pathway. However, they primarily by soluble inflammatory agents such as TNF-a, IFN-g, induce uncontrollable and fatal GvHD. IL-1, and NO. The attack on the epithelium, induced by the sol- Donor hematopoietic progenitors enter and restore the thymus, uble inflammatory agents, has been proved to not require Ag ex- and differentiate into T cells via the thymus-dependent pathway pression (30). This novel evidence explains the impairment of not after BMT. Because the recipient thymus is impaired by the only the recovery of the recipient thymus, but also the develop- conditioning regimen (especially irradiation), the thymic recon- ment of the transplanted thymus (Fig. 4B–E). Methods to inhibit stitution takes ∼10 wk; T cells are the last hematopoietic lineage the activated CD4+ T cell–induced inflammation, such as neu- cells to recover after BMT in mice and humans (32). To improve tralization of TNF-a and the administration of anti-CD4 Ab, by guest on September 26, 2021 this situation, we combined TT with BMT to provide a new en- ameliorated the effects of inflammatory impairment on the re- vironment for T cell induction and differentiation. TT was able to cipient epithelium or bone marrow niche (28, 30). Recent reports significantly enhance not only the percentage, but also the number have proved that FasL, which is a T cell cytolytic molecule, is of CD4+ T cells, which helped restore early T cell functions required for the thymus and bone marrow GvHD (22, 28). The whereas avoiding any concomitant GvHD (14, 27). results from one clinical study show a clear correlation between In this study, we combined IBM-BMT with TT and DLI, be- the level of serum-soluble FasL and liver damage during GvHD cause IBM-BMT can efficiently recruit donor-derived stromal (35). Based on these findings, we examined the concentration of cells (including mesenchymal stem cells), support the donor- TNF-a and soluble FasL in the plasma, and compared the levels in

FIGURE 6. The levels of TNF-a and FasL and percentage of Tregs. (A) The levels of TNF-a and FasL in the recipient plasma were measured 5 d after BMT. (B) Percentage of Tregs in the recipient splenocytes was analyzed 4 wk after BMT. The Tregs were of donor origin as they showed H-2Kb+. n = 3–4/group; experiments were performed three times. *p , 0.05, **p , 0.01. 2982 THYMUS IMPAIRMENT BY CD4+ T CELLS

2 the TT, WSP-DLI, and CD4 -DLI groups. The results showed a powerful new strategy for treatment of intractable autoimmune diseases in MRL/lpr mice. Blood 97: 3292–3299. that the TNF-a and FasL levels in the plasma in the WSP-DLI 8. Li, Q., H. Hisha, R. Yasumizu, T. X. Fan, G. X. Yang, Q. Li, Y. Z. Cui, group were much higher than in the other two groups. Blockade X. L. Wang, C. Y. Song, S. Okazaki, et al. 2007. Analyses of very early he- of TNF-a and FasL by Abs prevented inflammatory cytokine- mopoietic regeneration after bone marrow transplantation: comparison of in- travenous and intrabone marrow routes. Stem Cells 25: 1186–1194. mediated impairment of both recipient and transplanted thymuses, 9. Koike, Y., Y. Adachi, Y. Suzuki, M. Iwasaki, N. Koike-Kiriyama, K. Minamino, confirming the role of TNF-a and FasL in thymic impairment. In K. Nakano, H. Mukaide, A. Shigematsu, Y. Kiyozuka, et al. 2007. Allogeneic addition, the percentage of Tregs that could suppress the inflam- intrabone marrow-bone marrow transplantation plus donor lymphocyte infusion suppresses growth of colon cancer cells implanted in skin and liver of rats. Stem matory reaction was also decreased dramatically after WSP-DLI, Cells 25: 385–391. 2 but not CD4 -DLI. It thus appears that the inflammation induced 10. Ikehara, S. 2009. A new bone marrow transplantation method for stem cell by the allo-CD4+ T cells mediates the unselective impairment of disorders. Ann. N. Y. Acad. Sci. 1173: 774–780. 11. Mackall, C. L., T. A. Fleisher, M. R. Brown, M. P. Andrich, C. C. Chen, both the recovery of the recipient’s thymus and the development I. M. Feuerstein, M. E. Horowitz, I. T. Magrath, A. T. Shad, S. M. Steinberg, of the transplanted thymus. et al. 1995. Age, thymopoiesis, and CD4+ T-lymphocyte regeneration after in- tensive chemotherapy. N. Engl. J. Med. 332: 143–149. There are thymic epithelial cells, endothelial cells, and fibro- 12. Weinberg, K., G. Annett, A. Kashyap, C. Lenarsky, S. J. Forman, and blasts in the thymic stroma. The destruction on the thymic stroma R. Parkman. 1995. The effect of thymic function on immunocompetence fol- impairs the thymic cellularity and function. Recipient thymic lowing bone marrow transplantation. Biol. Blood Marrow Transplant. 1: 18–23. 13. Krenger, W., and G. A. Holla¨nder. 2008. The immunopathology of thymic GvHD is largely due to the impairment of the stromal cells of the GVHD. Semin. Immunopathol. 30: 439–456. thymus induced by allo-activated T cells. We have further proved 14. Markert, M. L., B. H. Devlin, M. J. Alexieff, J. Li, E. A. McCarthy, S. E. Gupton, a causative link between the activated CD4+ T cells (but not CD8+ I. K. Chinn, L. P. Hale, T. B. Kepler, M. He, et al. 2007. Review of 54 patients with complete DiGeorge anomaly enrolled in protocols for thymus transplan- T cells) and the impairment of the thymic stromal cells using an tation: outcome of 44 consecutive transplants. Blood 109: 4539–4547. Downloaded from in vitro system (data not shown). 15. Ikehara, S. 2011. Thymus transplantation for treatment of cancer: lessons from Newborn TT increased the percentage of CD4+ T cells signif- murine models. Expert Rev. Clin. Immunol. 7: 205–211. 16. Hosaka, N., T. Ryu, T. Miyake, W. Cui, T. Nishida, T. Takaki, M. Inaba, and icantly in the peripheral blood after BMT (27). This population S. Ikehara. 2007. Treatment of autoimmune diseases in MRL/lpr mice by allo- did not induce any serious recipient organ injury, suggesting that genic bone marrow transplantation plus adult thymus transplantation. Clin. Exp. Immunol. 147: 555–563. these cells experienced reconstitution via a thymus-dependent 17. Li, M., N. G. Abraham, L. Vanella, Y. Zhang, M. Inaba, N. Hosaka, S. Hoshino, pathway. We have obtained encouraging results using thymus- M. Shi, Y. M. Ambrosini, M. E. Gershwin, and S. Ikehara. 2010. Successful http://www.jimmunol.org/ independent CD8+ T cells in treating solid tumors (36). This modulation of type 2 diabetes in db/db mice with intra-bone marrow—bone 2 marrow transplantation plus concurrent thymic transplantation. J. Autoimmun. infers that the graft-versus-tumor effects of TT and CD4 -DLI can 35: 414–423. be combined for tumor suppression without severe thymic im- 18. Zhang, Y., N. Hosaka, Y. Cui, M. Shi, and S. Ikehara. 2011. Effects of allogeneic pairment or GvHD. hematopoietic stem cell transplantation plus thymus transplantation on malig- nant tumors: comparison between fetal, newborn, and adult mice. Stem Cells In this study, we have shown that DLI leads to derangements in Dev. 20: 599–607. both recipient and transplanted thymuses in the IBM-BMT + TT 19. Tomblyn, M., and H. M. Lazarus. 2008. Donor lymphocyte infusions: the long and setting. The inflammation mediated by allo-CD4+ T cells plays winding road: how should it be traveled? Bone Marrow Transplant. 42: 569–579. 20. Sykes, M. 2009. Hematopoietic cell transplantation for tolerance induction: a central role in the impairment of the recipient thymic recovery animal models to clinical trials. Transplantation 87: 309–316. and the transplanted thymic development. The impairment would 21. Frey, N. V., and D. L. Porter. 2008. Graft-versus-host disease after donor leu- by guest on September 26, 2021 be significantly ameliorated by the depletion of CD4+ T cells. kocyte infusions: presentation and management. Best Pract. Res. Clin. Hae- matol. 21: 205–222. 22. Hauri-Hohl, M. M., M. P. Keller, J. Gill, K. Hafen, E. Pachlatko, T. Boulay, Acknowledgments A. Peter, G. A. Holla¨nder, and W. Krenger. 2007. Donor T-cell alloreactivity against host thymic epithelium limits T-cell development after bone marrow We thank Hilary Eastwick-Field and Keiko Ando for help in the prepa- transplantation. Blood 109: 4080–4088. ration of this manuscript. We thank Aiko Kitajima for excellent technical 23. Na, I. K., S. X. Lu, N. L. Yim, G. L. Goldberg, J. Tsai, U. Rao, O. M. Smith, assistance regarding morphology. C. G. King, D. Suh, D. Hirschhorn-Cymerman, et al. 2010. The cytolytic mol- ecules Fas ligand and TRAIL are required for murine thymic graft-versus-host disease. J. Clin. Invest. 120: 343–356. Disclosures 24. Shi, M., Y. Adachi, A. Shigematsu, N. Koike-Kiriyama, W. Feng, S. Yanai, C. Yunze, Z. X. Lian, J. Li, and S. Ikehara. 2008. Intra-bone marrow injection of The authors have no financial conflicts of interest. donor bone marrow cells suspended in collagen gel retains injected cells in bone marrow, resulting in rapid hemopoietic recovery in mice. Stem Cells 26: 2211–2216. 25. Shi, M., Y. Adachi, Y. Cui, M. Li, Z. Lian, Y. Zhang, S. Yanai, C. Shima, Y. Imai, References and S. Ikehara. 2011. Combination of intra-bone marrow-bone marrow trans- 1. Friedrich, W. 1996. Bone marrow transplantation in immunodeficiency diseases. plantation and subcutaneous donor splenocyte injection diminishes risk of graft- Ann. Med. 28: 115–119. versus-host disease and enhances survival rate. Stem Cells Dev. 20: 759–768. 2. Eibl, B., H. Schwaighofer, D. Nachbaur, C. Marth, A. Ga¨chter, R. Knapp, 26. Hattori, K., T. Hirano, H. Miyajima, N. Yamakawa, M. Tateno, K. Oshimi, N. Kayagaki, H. Yagita, and K. Okumura. 1998. Differential effects of anti-Fas G. Bo¨ck, C. Gassner, L. Schiller, F. Petersen, and D. Niederwieser. 1996. Evi- ligand and anti-tumor necrosis factor alpha antibodies on acute graft-versus-host dence for a graft-versus-tumor effect in a patient treated with marrow ablative disease pathologies. Blood 91: 4051–4055. chemotherapy and allogeneic bone marrow transplantation for breast cancer. 27. Ryu, T., N. Hosaka, T. Miyake, W. Cui, T. Nishida, T. Takaki, M. Li, Blood 88: 1501–1508. K. Kawamoto, and S. Ikehara. 2008. Transplantation of newborn thymus plus 3. Burt, R. K., A. E. Traynor, R. Pope, J. Schroeder, B. Cohen, K. H. Karlin, hematopoietic stem cells can rescue supralethally irradiated mice. Bone Marrow L. Lobeck, C. Goolsby, P. Rowlings, F. A. Davis, et al. 1998. Treatment of Transplant. 41: 659–666. autoimmune disease by intense immunosuppressive conditioning and autologous 28. Shono, Y., S. Ueha, Y. Wang, J. Abe, M. Kurachi, Y. Matsuno, T. Sugiyama, hematopoietic stem cell transplantation. Blood 92: 3505–3514. T. Nagasawa, M. Imamura, and K. Matsushima. 2010. Bone marrow graft- 4. Bregni, M., A. Dodero, J. Peccatori, A. Pescarollo, M. Bernardi, I. Sassi, versus-host disease: early destruction of hematopoietic niche after MHC- C. Voena, A. Zaniboni, C. Bordignon, and P. Corradini. 2002. Nonmyeloablative mismatched hematopoietic stem cell transplantation. Blood 115: 5401–5411. conditioning followed by hematopoietic cell allografting and donor lymphocyte 29. Ferrara, J. L., J. E. Levine, P. Reddy, and E. Holler. 2009. Graft-versus-host infusions for patients with metastatic renal and breast cancer. Blood 99: 4234– disease. Lancet 373: 1550–1561. 4236. 30. Teshima, T., R. Ordemann, P. Reddy, S. Gagin, C. Liu, K. R. Cooke, and 5. Marmont, A. M., F. Gualandi, G. Piaggio, M. Podesta`, M. Teresa van Lint, J. L. Ferrara. 2002. Acute graft-versus-host disease does not require alloantigen A. Bacigalupo, and F. Nobili. 2006. Allogeneic bone marrow transplantation expression on host epithelium. Nat. Med. 8: 575–581. (BMT) for refractory Behc¸et’s disease with severe CNS involvement. Bone 31. Krenger, W., and G. A. Holla¨nder. 2008. The thymus in GVHD pathophysiology. Marrow Transplant. 37: 1061–1063. Best Pract. Res. Clin. Haematol. 21: 119–128. 6. Ikehara, S. 2011. A novel BMT technique for treatment of various currently 32. Zlotoff, D. A., S. L. Zhang, M. E. De Obaldia, P. R. Hess, S. P. Todd, intractable diseases. Best Pract. Res. Clin. Haematol. 24: 477–483. T. D. Logan, and A. Bhandoola. 2011. Delivery of progenitors to the thymus 7. Kushida, T., M. Inaba, H. Hisha, N. Ichioka, T. Esumi, R. Ogawa, H. Iida, and limits T-lineage reconstitution after bone marrow transplantation. Blood 118: S. Ikehara. 2001. Intra-bone marrow injection of allogeneic bone marrow cells: 1962–1970. The Journal of Immunology 2983

33. Zhang, Y., J. P. Louboutin, J. Zhu, A. J. Rivera, and S. G. Emerson. 2002. 35. Das, H., S. Imoto, T. Murayama, K. Kajimoto, T. Sugimoto, T. Isobe, Preterminal host dendritic cells in irradiated mice prime CD8+ T cell-mediated T. Nakagawa, R. Nishimura, and T. Koizumi. 1999. Levels of soluble FasL and acute graft-versus-host disease. J. Clin. Invest. 109: 1335–1344. FasL gene expression during the development of graft-versus-host disease in 34. Schmaltz, C., O. Alpdogan, K. J. Horndasch, S. J. Muriglan, B. J. Kappel, DLT-treated patients. Br.J.Haematol.104: 795–800. T. Teshima, J. L. Ferrara, S. J. Burakoff, and M. R. van den Brink. 2001. Dif- 36. Suzuki, Y., Y. Adachi, K. Minamino, Y. Zhang, M. Iwasaki, K. Nakano, ferential use of Fas ligand and perforin cytotoxic pathways by donor T cells in Y. Koike, and S. Ikehara. 2005. A new strategy for treatment of malignant tumor: graft-versus-host disease and graft-versus-leukemia effect. Blood 97: 2886– intra-bone marrow-bone marrow transplantation plus CD4- donor lymphocyte 2895. infusion. Stem Cells 23: 365–370. Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021