1、 在小鼠骨髓移植中 CCR5 与急性移植物抗宿主病的相关性研究【摘要】 本研究评价供者 CCR5 在经过强化预处理的骨髓移植动物模型受者体内的作用，为今后的异基因造血干细胞移植的临床应用提供科学依据。经过致死剂量照射的 BALB/c 小鼠接受异基因C57BL/6 小鼠的骨髓移植。根据回输的细胞不同实验分为 4 组：B6 CCR5 KO 组，受者接受 C57BL/6 CCR5-/-小鼠骨髓和脾脏细胞；B6 WT 组，受者接受野生型 C57BL/6 小鼠骨髓和脾脏细胞；B6 CCR5 KO BMC 组，受者只接受 C57BL/6 CCR5-/-小鼠骨髓细胞；B6 WT BMC 组，受者只接受野生型
2、 C57BL/6 小鼠骨髓细胞。结果表明：较之 B6 WT 组，B6 CCR5 KO 组小鼠以更快的速度死于急性GVHD；其受者体内的 CD8+T 细胞更大量的增殖；其 T 细胞恢复后产生更多的 INF- 和 TNF- 并且由于其 T 细胞有丝分裂原刀豆素水平处于较高水平，从而进一步促进 T 细胞的增殖，提示 CCR5 的作用之一是下调参与排异反应的供者 CD8+T 细胞的增殖。组织学评价提示，移植剔除 CCR5 基因受者细胞的小鼠肾脏出现了病理损伤并且肝脏存在有更为严重的病理变化。结论：剔除 CCR5 基因的异基因骨髓移植使 GVHD 发病率的增加，供者 CD8+T 细胞在受者体内增殖增加以
3、及肝肾损害加重，这提示 CCR5 在异基因骨髓移植中起着重要作用。 【关键词】 CCR5； GVHD； 异基因造血干细胞移植； CD8+细胞Graft-vs-host disease (GVHD) remains a major source of morbidity in patients after allogeneic bone marrow transplantation (BMT). Acute GVHD is a complex pathological process involving numerous cell types and target tissues. At its
4、 core, acute GVHD starts out as an immunological recognition of the donor T cells to the genetically disparate recipient. However, the process then amplifies as these T cells expand and, with the help of inflammatory cytokines (in part produced in response to the intensive condi-tioning in BMT as we
5、ll as the T cells themselves), recruit other cells in the generation of organ damage1,2.This in part is the“cytokine storm” that fuels GVHD pathogenesis3. Multiple organs are affected in acute GVHD including the skin, gut, lungs, and liver. While many factors can influence the generation of GVHD, it
6、 is clear that the extensive conditioning in BMT contributes to the generation and pathology of GVHD in both man and mouse4,5, emphasizing the importance of models that can reflect the conditioning regimens used in clinical BMT. Chemokines are a diverse group of cytokines that modulate immune cell t
7、rafficking and function6,7. Due to the properties of chemokines, it has been postulated that chemokines may play a role in GVHD. Serody et al8 demonstrated reduced lethality in recipients of donor T cells from MIP-1-deficient mice compared to wild-type (WT) T cells in a CD8+ T cell-mediated GVHD mod
8、el. Recent reports also indicated that blockade of CCR5 using neutralizing antibodies could also protect from GVHD using a parent into F1 GVHD model9 . This report and another using CCR5 KO mice in the same model indicated that CCR5 on cells played a role in augmenting the GVHD response and that blo
9、ckade of this interaction could be used to prevent GVHD9,10. However, the precise mechanism underlying this protection was not elucidated. Interestingly, there have also been reports that CCR5 may play a role in down regulating T cell function as evidenced by some infectious disease models11, sugges
10、ting that this receptor-ligand interaction may exert diverse effects. The previous reports assessing the role of CCR5 expression on GVHD used a model in which no conditioning was administered. Chemokine production as well as receptor expression and function have been shown to be markedly influenced
11、by inflammatory cytokines induced after total-body irradiation (TBI)12,13. We sought to assess the role of CCR5 on donor cells in models where extensive conditioning of the recipient was applied, thus mimicking the regimens used in conventional clinical BMT. Our results demonstrate that the absence
12、of CCR5 on donor cells results in a significant increase in GVHD morbidity that was associated with an increase in donor CD8+ T-cell expansion in the recipients.Materials and MethodsMiceC57BL/6(B6; H2b)and BALB/c(H2d)were purchased from the Animal Production Area (NCI at Frederick,Frederick,MD,USA).
13、 C57BL/6 CCR5-/- mice (B6 CCR5 KO) were provided from two separate colonies by Dr. Jonathan Serody (Univ. North Carolina School of Medicine, Chapel Hill, NC, USA) and Dr. William Kuziel. All recipients were age-matched females and were 2 to 6 months of age at the time of BMT.Cell preparation Bone ma
14、rrow cell (BMC) suspensions were prepared by gently releasing cells from the backbones, femurs, and tibiae into Dulbeccos phosphate-buffered saline solution (DPBS) with a mortar and pestle, filtering through amesh filter to remove particulates, and washing the cell suspensions twice. Spleen and lymp
15、h node cell preparations were prepared by gently crushing the tissues to release the cells. Preparations were filtered to remove debris and washed twice in DPBS for injection or RPMI 1640 supplemented with 5% fetal bovine serum (FBS) for in vitro assays. Cell counts were performed on a Coulter Z1 ce
16、ll counter (Coulter Electronics, Hialeah, FL, USA).Induction of GVHDFor bone marrow transplants(BMT),recipient BALB/c mice received lethal total-body irradiation of 800 cGy by a 137Cesium source. The mice then received 107 bone marrow cells (BMC) and （7-8）106 splenocytes and lymph node cells from do
17、nor B6 or B6 CCR5 KO mice. Mice were monitored and weighed weekly. All moribund mice were euthanized.Flow cytometryT cell constituency and donor origin of the splenocytes were measured using CD4, CD8, CD3, and H2Db mAbs, respectively, obtained from PharMingen (San Diego,CA,USA).Three-color flow cyto
18、metry, using fluorescein isothiocyanate (FITC), phycoerythrin (PE), or biotin- (along with streptavidin-cychrome) conjugated mAb purchased from PharMingen, was performed. Irrelevant mAb control values were subtracted from values obtained with relevant mAbs. All results were obtained using a FACScan
19、(Becton-Dickinson, San Diego,CA,USA).Forward- and side-scatter settings were gated to exclude red cells and debris. Ten thousand cells were analyzed for each determination.Histological evaluatianTissues from the mice were placed in 10% formalin, imbedded in paraffin, sectioned, and stained with hema
20、toxylin and eosin. Tissue sections were evaluated and graded blind by a veterinary pathologist (MRA) as previously described14. A semi-quantitative scale from 0 to 4 was used where histopathological changes were identified as minimal=1, mild=2, moderate=3, severe=4. A total of 3 to 10 mice per exper
21、imental group combined from up to four independent experiments were assessed.Lymphokine productionUnfractionated spleen cells from 7 days after BMT were plated in triplicate 96-well plates at 1 million cells per ml in RPMI 1640 (with 10% FBS, 2 mmol/L glutamine, 510-5mol/L 2-mercaptoethanol, 100 U/m
22、l penicillin, and 100 g/ml streptomycin with or without 10 g/ml Con A (Sigma Chemical Co.,St.Louis,MO, USA). Cells were incubated at 37 with 5% CO2 in a humidified incubator. On day 3 of culture, the number of viable cells per well were determined by trypan blue exclusion. Cell-free supernatants wer
23、e collected and frozen at 20. Frozen supernatants were thawed and assayed for cytokine production according to manufacturers specifications. ELISA kits were purchased from RD Systems (Minneapolis, MN, USA) for determination of mouse interferon-(IFN-) and tumor necrosis factor- (TNF-).Statistical ana
24、lysisSurvival data were plotted by the Kaplan-Meier method and analyzed by the log-rank test. Comparisons of cell populations and lymphokine production were made with Students t-test. Pathology scores were evaluated with the Fishers exact test. A P value of equal to or less than 0.05 was considered
25、as significant.ResultsIncreased acute GVHD in recipients receiving cells from CCR5 KO mice after allogeneic BMTThe effects of CCR5 KO donor cells in a model of GVHD was first to be assessed, where the recipients received lethal TBI followed by a full MHC-mismatched allogeneic BMT. Recipient BALB/c (
26、H2d) mice received lethal TBI (800 cGy) followed by donor cell infusion from CCR5 KO and wild-type control mice on C57BL/6 background. Both BMC and donor T cells were transferred into the irradiated recipients in order to induce GVHD and recipients receiving BMC alone were used as controls. Flow cyt
27、ometric assessment indicated that both CCR5 KO mice and their wildtype controls had similar numbers of T cells in their spleens (Table 1), indicating that comparable numbers of T cells were administered in these models. GVHD was assessed by monitoring progressive weight loss with subsequent histolog
28、ical assessment of key target organs (skin, gut, lungs, kidney, and liver). Surprisingly, the results demonstrated that use of CCR5 KO donor cells resulted in significant (P0.001) increases in GVHD morbidity compared to mice receiving wild-type cells (Figure 1). These results suggest that the loss o
29、f CCR5 on donor cells results in increased GVHD after allogeneic BMT in models using extensive conditioning.Table 1.T cell subsets in C57BL/6 wild-type and CCR5 KO mice (略)As CCR5 KO mice have been reported to have increased CD8+ T-cell responses in some infectious disease models11, we then assessed
30、 the extent of donor cell engraftment and function in the recipients undergoing GVHD. We hypothesized that CCR5 on the donor T cells may play a role in limiting T-cell function and thus increases in donor T cell engraftment and activity may be detected in the mice, which would correlate with the inc
31、reased GVHD morbidity associated with the use of CCR5 KO donors. BALB/c mice received lethal TBI followed by donor cells from either CCR5 KO or control wild-type recipients. Seven days after BMT, before any mice had succumbed to GVHD, spleens were removed from some mice and assessed for donor cell e
32、ngraftment. The results demonstrate that the mice receiving cells from CCR5 KO donors had significantly greater donor CD8+ T-cell engraftment in the spleens when compared to recipients of wild-type donors(P0.01)(Figure 2A).Donor CD4+ T-cell engraftment was not significantly different although slight
33、ly increased as well(Figure 2B).The donor CD8+T cells from the CCR5 KO donorsFigure 2. Increased numbers of splenic CD8+T cells in CCR5 KO cell recipients.BALB/c recipient mice received B6 wild-type or CCR5 KO bone marrow (107 cells) and (7-8) 106 splenocytes and lymph node cells after myeloabative
34、TBI. On day 7 post-BMT significant increases (Students t test; P0.05) in donor (H2Db) CD8+T cells and donor CD8+CD25+ cells but not donor-derived CD4+ T cells were observed. Three mice per group were analyzed for the donor-derived T cell content in spleen by flow cytometry.also were significantly gr
35、eater in their expression of CD25 (P0.05), indicating that they were in an activated state. When spleen cells from these mice were then cultured for 72 hours, it was found that splenocytes from the mice receiving cells from the CCR5 KO donors produced significantly(P0.01) greater amounts of TNF- and
36、 IFN-(Figure 3A). Splenocytes from these mice also proliferated to the T cell mitogen, concanavalin A,to a significantly(P0.01)greater extent when compared to recipients receiving cells from recipients of wild-type donors (Figure 3B).Taken together,these results indicate that after allogeneic BMT,us
37、e of donor cells from CCR5 KO mice resulted in significantly greater donor CD8+ T-cell expansion and function as reflected by proliferation and cytokine production when compared to recipients of wild-type donors. Furthermore, this increase in CD8+ T cell expansion could be correlated with increased
38、morbidity from acute GVHD observed in these recipients. Figure 3. Enhanced production of TNF- and IFN- in cultures of splenocytes taken from day 7 post-BMT. One million splenocytes from day 7 post-BMT BALB/c recipients of B6 WT or B6 CCR5 KO hematopoietic grafts were cultured in media alone or with
39、10 g/ml ConA for 3 days. Viable cells were assessed on a hemacytometer and trypan blue exclusion. Supernatants were collected and analyzed by ELISA for lymphokine production (3 mice per group).Increased liver pathological changes in mice undergoing acute GVHD receiving cells from CCR5 KO miceAs oppo
40、sed to other models of GVHD in which no conditioning is applied, acute GVHD in models where extensive conditioning is applied affects multiple target organs: skin, gut, lungs, and liver. It is believed that the presence of prior conditioning not only predisposes these tissues to further damage by th
41、e donor cells, but the conditioning induces potent inflammatory cytokines that add to the “cytokine storm” associated with acute GVHD4,5. Examination of target visceral organs for GVHD revealed diffuse severe hepatocellular vacuolation in the livers of recipients of CCR5 KO donor T cells (Table 1) w
42、hen compared to mice receiving donor T cells from wild-type controls. Mice undergoing increased acute GVHD after receiving CCR5 KO donor cells also had renal lesions (Table 1), cortical tabular vacuolation, tubular cell necrosis, and cellular/proteinaceous casts. Renal damage did not occur in the re
43、cipients of WT grafts or CCR5 KO bone marrow grafts. These results would suggest that the accelerated morbidity from acute GVHD could be correlated with increases in liver and kidney lesions among the recipients receiving the CCR5 KO donor T cells. The intestinal tract is a major target for GVHD. In
44、 this model,we did not observe any consistent differ- ences in the degree of GVHD-associated pathological changes in the small or large intestine between recipients of WT or CCR5 KO donor cells (Table 2). No appreciable differences in the extent of tissue damage were detected in the skin, lung, or o
45、ther organs among the recipients. Thus, increases in GVHD morbidity associated with the use of CCR5 KO donor T cells could be correlated with increases in liver and kidney lesions among the recipients.DiscussionSimilar to cytokines, the role of chemokines in GVHD are most likely complex and highly d
46、ependent on the model used to assess them. This report demonstrated that the absence of CCR5 on donor cells resulted in accelerated morbidity associated with GVHD and heavier damage in two target organs, the liver and the kidney. The absence of CCR5 on the donor cells resulted in a net acceleration
47、of acute GVHD with increased donor CD8+T-cell expansion. This is in agreement with a recent report by Serody et al that also demonstrated increased GVHD when T cells from CCR5 KO mice were used as donors in models using conditioning15. It is important to note that the word “extensive” is relative. S
48、ome GVHD models use considerably higher amounts of conditioning in the study of GVHD pathology, particularly related to the gut and its breakdown5. The heavier GVHD observed with CCR5 KO donors seems to be contradictory to reports demonstrating that antibodies to CCR5 or using CCR5 KO mice were prot
49、ective to GVHD9,10. The discrepancy may be due to differences in the models and means to deprive CCR5 interactions. In the aforementioned model, large numbers of parental cells were transferred into F1 recipients and no conditioning was administered to the recipients9,10. This GVHD model has shown to depend largely on the attack of the host lymphoid cells by the donor CD8+ T cells in a one-way immune reaction16. The role of conditioning in BMT models has been previously reported to have dramatic effects on the role of cytokines in the p
