GVHD after chemotherapy conditioning in allogeneic transplanted mice

ORIGINAL ARTICLE

GVHD after chemotherapy conditioning in allogeneic transplanted mice

B Sadeghi1, N Aghdami1, Z Hassan1,2, M Forouzanfar1, B Rozell3, M Abedi-Valugerdi4

and M Hassan1,5

1Experimental Cancer Medicine, Institution for Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden; 2Center forAllogeneic Stem Cell Transplantation (CAST), Karolinska University Hospital-Huddinge, Stockholm, Sweden; 3Morphology andPhenotype analysis, Institution for Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden; 4Department of Biochemistryand Biophysics, Arrhenius Laboratories for the Natural Sciences, Stockholm University, Stockholm, Sweden and 5Clinical ResearchCenter, Karolinska University Hospital, Huddinge, Stockholm, Sweden

GVHD is a major complication in allogeneic SCT.Available GVHD models are mainly based on radio-therapy-conditioning and/or immune deficient mice.GVHD models based on chemotherapy-based regimensremain poorly studied, despite 50% of all transplantationsbeing chemotherapy based. Our aim was to develop aGVHD model using chemotherapy as conditioning.Female BALB/c (H-2Kd) were conditioned with BU–CYand transplanted with 2� 107 BM and 3� 107 spleen cellsfrom either C57BL/6 (H-2Kb) mice (allogeneic setting)or from male BALB/c to serve as a control group forregimen-related toxicity and engraftment. GVHD mani-festations and histopathological changes were evaluated.Chimerism and donor T cells presence in skin, intestineand liver were studied using FACS-, FISH analysis andimmunohistochemistry. Allogeneic transplanted mice de-veloped lethal GVHD starting from dayþ 7 with bothhistological and clinical signs. Donor T cells accumulatedin recipient skin and intestine with GVHD progression.BM-failure, apoptosis and T-lymphocyte infiltration intotarget organs were significantly higher in allogeneic whencompared with the syngeneic group. No toxicity orGVHD signs were observed in the syngeneic setting. Wereport a mouse model of GVHD using BU–CY condition-ing that represents the most common myeloablative-conditioning regimen in clinical SCT. This model can beutilized to study the role of conditioning on mechanismsunderlying GVHD.Bone Marrow Transplantation (2008) 42, 807–818;doi:10.1038/bmt.2008.261; published online 29 September 2008Keywords: BU; CY; chemotherapy conditioning; GVHD;mouse model; hematopoietic stem cell transplantation

Introduction

Hematopoietic stem cell transplantation (HSCT) is acurative therapy for a wide variety of diseases, includinghematological disorders,1 metabolic diseases, immunedeficiencies, autoimmune diseases and solid tumors.2

However, the success of HSCT is limited by severalcomplications such as acute GVHD (aGVHD),3 graftrejection, veno-occlusive disease, infections and relapse.4,5

Mechanisms underlying the development of aGVHD arenot fully understood. However, it is known that aGVHDemerges when donor T cells are activated by the hostantigen-presenting cells, thereby inducing antigen-specificeffector T cells.6 The effector cells express high levels oftissue-specific adhesion and chemoattractant receptors,facilitating migration into target epithelial organs such asskin, intestine, lungs and liver.7 The infiltration of effectorcells into organs, is usually accompanied by non-specificrelease of pro-inflammatory cytokines for example, tumornecrosis factor, IL-1, IL-2 and IL-8, that have deleteriouseffects in the affected organs.8 Apart from the histo-compatibility disparities between donor and recipient,pre-transplant conditioning plays a crucial role in thedevelopment and severity of GVHD.9,10

Conditioning11 has multiple functions in the recipient,including; the depletion of stem cells,12 suppression of thehost immune system and most importantly eliminationof tumor cells.13 A combination of CY and TBI wasintroduced into the clinic as a conditioning regimen about30 years ago.1,14 During the early 1980s, an irradiation-freeconditioning regimen consisting of BU and CY wasintroduced.15 Comparative studies have shown that bothconditioning regimens exhibit comparable potencies in theirantileukemic and immunosuppressive effects. Furthermore,comparing both regimens showed no significant differencesin the engraftment rates, survival, relapse and transplant-ation-related mortality.16,17

Animal models of HSCT are widely used to elucidatemechanisms underlying the development of aGVHD andfurther to provide strategies for its treatment. The majorityof animal models of HSCT and GVHD are based on TBIas a conditioning regimen. On account of the variety ofconditioning used in the clinical settings and because of the

Received 16 June 2008; revised 3 July 2008; accepted 17 July 2008;published online 29 September 2008

Correspondence: Dr M Hassan, Experimental Cancer Medicine,Laboratory Medicine, Clinical Research Center, Karolinska UniversityHospital, Huddinge, Stockholm 141 86, Sweden.E-mail: [email protected]

Bone Marrow Transplantation (2008) 42, 807–818& 2008 Macmillan Publishers Limited All rights reserved 0268-3369/08 $32.00

www.nature.com/bmt

http://dx.doi.org/10.1038/bmt.2008.261

mailto:[email protected]

http://www.nature.com/bmt

different mechanism of action of TBI and chemotherapy,available aGVHD models cannot cover different aspects ofaGVHD in clinical settings since over 50% of the patientsare conditioned with chemotherapy. There is an essentialneed for an animal model of aGVHD that mimics theclinical setting. Thus, in the present investigation, weintroduce a mouse model of aGVHD where BU–CY wasused as the conditioning regimen. The transplanted animalsexhibited clinical and pathological manifestations, whichcompletely fulfilled the criteria for aGVHD. The presentmodel is reliable and can be utilized to elucidate theunderlying mechanism for aGVHD developed after che-motherapy conditioning in allogeneic transplanted miceand to evaluate investigational therapies.

Materials and methods

AnimalsFemale BALB/c (H-2Kd), female and male C57BL/6 (H-2Kb) mice (10–12 weeks old) were purchased from Scanbur(Sollentuna, Sweden). Mice were maintained under patho-gen-free controlled conditions and 12-hr light/darkness.Animals had access to food and water ad libitum.

Dose optimizationBU and CY were obtained from Sigma (Stockholm,Sweden). BU was dissolved in DMSO (40mg/ml) and CYin sterile water (10mg/ml). Seven groups of female BALB/cand one group of C57BL/6 received IP doses of BU (20 or25mg/kg/day) for 4 days, followed by CY (100 or 150mg/kg/day) for 2 days. Days �1 and 0 were the resting andBMT days, respectively. Control group (BALB/c) wasinjected i.p. with vehicles only. The conditioning protocoland groups are summarized in Table 1. All experimentsdescribed here were approved by the Ethical Committee forAnimal Experimentation.

BMTOn day 0, donor male C57BL/6 and female BALB/c micewere killed by cervical dislocation. Bone marrow cells(BMC) were flushed from femur and tibia and single-cellsuspension was prepared in RPMI-1640 (RPMI) containingFBS. Cells were then centrifuged and washed with RPMI.Donor spleens were disrupted in RPMI-containing FBS toform cell suspension. Splenocytes were passed through a70 mm strainer, centrifuged and washed two times withRPMI. Recipient mice were injected i.v. with 20� 106 BMCcombined with 30� 106 splenocytes from donors (Table 1).

Assessment of GVHDRecipient mice were monitored daily from conditioningstart until the appropriate sampling day for the clinicalmanifestations of GVHD (for example, weight loss,hunched posture, poor activity, ruffled fur and loss of skinintegrity).18 The severity of each given symptom was scoredfrom 0 to 2. The sum of the scores for all symptoms in eachmouse (maximally 10) was used as an index of the severityand progression of GVHD. Each experiment was repeated4–5 independent times.

Preparation of BM and spleen samplesRecipient mice were killed at appropriate time points, andspleens and femurs were removed. Single cell suspensions ofsplenocytes and BMC’s were prepared. Erythrocytes werelysed with ammonium chloride. The obtained single-cellsuspensions were washed two times in PBS and re-suspended in FACS buffer solution.

Isolation of epidermal immune cellsEpidermal immune cells in the skin were isolated asdescribed earlier.19 Briefly, pieces of skin (abdomen) wereplaced in a micro-tube containing 1ml Dispase andincubated overnight at 4 1C. Epidermis was removed andthe tissues were cut into small pieces. The samples were

Table 1 Definition of study groups

Experimentalgroups

Conditioning regimens Resting day Graft day Recipient Donor Graft content No. of mice

Group 1 BU (80mg/kg) +CY (200mg/kg) Day 1 Day 0 fBALB/c 2 mC57BL/63 BM cells (20� 106) 36Spleen cells (30� 106)

Days (�7 to �4) Days (�3 to �2)Group 2 BU (80mg/kg) +CY (300mg/kg) Day 1 Day 0 fBALB/c mC57BL/6 BM cells (20� 106) 17

Days (�7 to �4) Days (�3 to �2) Spleen cells (30� 106)Group 3 BU (100mg/kg) +CY (200mg/kg) Day 1 Day 0 fBALB/c mC57BL/6 BM cells (20� 106) 17

Days (�7 to �4) Days (�3 to �2) Spleen cells (30� 106)Group 4 BU (100mg/kg) +CY (300mg/kg) Day 1 Day 0 fBALB/c mC57BL/6 BM cells (20� 106) 17

Days (�7 to �4) Days (�3 to �2) Spleen cells (30� 106)Group 5 BU (80mg/kg) +CY (200mg/kg) Day 1 Day 0 fBALB/c fBALB/c3 BM cells (20� 106) 20

Days (�7 to �4) Days (�3 to �2) Spleen cells (30� 106)Group 6 BU (100mg/kg) +CY (200mg/kg) Day 1 Day 0 fBALB/c fBALB/c BM cells (20� 106) 6

Days (�7 to �4) Days (�3 to �2) Spleen cells (30� 106)Group 7 BU (80mg/kg) +CY (200mg/kg) Day 1 Day 0 fC57BL/6 mC57BL/6 BM cells (20� 106) 13

Days (�7 to �4) Days (�3 to �2) Spleen cells (30� 106)Group 8 BU (80mg/kg) +CY (200mg/kg) — — fC57BL/6 — — 10

Days (�7 to �4) Days (�3 to �2)Group 9 (Control) Vehicle (DMSO) — — fC57BL/6 — — 15

Days (�7 to �2)

Mouse model of GVHD based on chemotherapyB Sadeghi et al

808

Bone Marrow Transplantation

then digested in Dulbecco’s modified Eagle’s mediumcontaining HEPES, DNase, collagenase D and hyaluroni-dase at 37 1C for 1.5 h. Thereafter, they were filteredthrough a 70 mm strainer to obtain single-cell suspensions.

Isolation of small intestine intraepithelial lymphocytesIntestinal lymphocytes (intraepithelial) were isolated asdescribed elsewhere.20 Briefly, the small intestine wasdissected and the lumen washed with cold PBS-2% FBS.The intestine was cut and placed in an Erlenmeyer flaskcontaining PBS–FBS–EDTA. The flask was incubated for20min at 37 1C with stirring. After 15 s of vigorous vortexthe supernatant was filtered and collected.

Preparation of blood samples for flow cytometryAt different time points, peripheral blood samples (80ml)were taken from a small incision in the tail and collected intubes containing 200ml PBS–EDTA. Blood was incubatedwith FC-blocker for 15min at 4 1C and then stained witha panel of fluorochrome-conjugated mAbs. Erythrocyteswere lysed and the stained cells were washed two times withFACS buffer and analyzed using FACSCalibur (BDSystems, San Jose, CA, USA).

Immunofluorescent staining and FACS analysisSplenocytes, BMC, skin and intestinal immune cells werestained using the following mAbs: anti-H-2Kb (clone: AF6-88.5-FITC), anti-H-2Kd (clone: SF1-1.1-FITC), anti-CD3(clone: AF6-88.5-FITC), anti-NK (clone: DX5-FITC), anti-H-2Kd (clone: SF1-1.1-PE), anti-CD8 (clone: 53-6.7-PE),anti-H-2Kb (clone: AF6-88.5-PE), anti-CD3 (clone: 145-2C11-PerCp), anti-CD25 (clone: PC61-PerCp), anti-CD11b(clone: M1/70-PerCp), anti-CD4 (clone: RM4-5-APC), anti-CD19 (clone: 1D3-APC) and anti-CD11c (clone: HL3-APC),all were purchased from (BD Biosciences, San Jose, CA,USA). For immunophenotyping, cells were first incubatedwith an FC-receptor blocking monoclonal antibody (clone:2.4G2) for 15min at 4 1C and then directly stained with apanel of mAbs for 30min at 4 1C. Finally, the stained cellswere washed two times with FACS buffer solution andanalyzed on a FACSCalibur flow cytometer.

Analysis and detection of chimerism using fluorescencein situ hybridization (FISH)At days þ 7, þ 14, þ 21, þ 90 and þ 120, BM and spleencell suspensions were applied onto slides using cyto-centrifugation (25 000–50 000 cells/slide). FISH analysiswas performed to detect the Y-chromosome as describedearlier.21 Briefly the cell nucleus was digested in HCl andpepsin. The probe, a total paint CY3-labeled mousechromosome-Y DNA, was applied to the target area thatsimultaneously was denatured in a HYBrite chamber. Theslides were washed, air-dried and counterstained withVectaShield DAPI (40,6-diamidine-2-phenylidole-dihy-drochloride)/antifade. Signal spot counting of Y-chromo-some was performed using a Nikon Eclipse-E800fluorescence microscope. For documentation, an OlympusBH-2 microscope connected to CytoVision image system(Applied Imaging International Ltd, Newcastle-upon,Tyre, UK) was used.

Histology and ImmunohistochemistryTissue samples were fixed in formalin for 24-h, transferredto 70% ethanol, dehydrated and embedded in paraffin.Immunohistochemical detection of CD3epsilon and cleavedCaspase-3 was performed using rabbit monoclonal anti-bodies, SP7 (Labvision, Fremont, CA, USA) and 5A1 (CellSignalling Technology, Danvers, MA, USA), respectively.Briefly 4–5mm sections were deparaffinized. Antigen retrievalwas performed in a 2100 retriever and Buffer-A according tothe manufacturer’s instructions. Slides were rinsed, treatedwith H2O2 in methanol and blocked by goat serum in Tris-buffer. Primary antibodies were diluted in the blockingsolution and applied at 4 1C overnight. After rinsing inTris-buffer, a biotin-labelled secondary anti-rabbit antibodywas applied. Sections were incubated with ABC–HRPcomplex (DakoCytomation, Glostrup, Denmark). Bindingsites were visualized with diaminobenzidine/hydrogen per-oxide, and the slides finally counterstained with hematoxylin.

Statistical analysisAll data are expressed as means±s.e. (standard error)unless otherwise mentioned. Differences between allogeneicand syngeneic analyzed using Mann–Whitney (U-test).Po0.05 is considered statistically significant. Survivalcurves were plotted using Kaplan–Meier estimates. Allstatistical analyses were performed utilizing SPSS ver13.

Results

Dose optimization and clinical manifestations of GVHDFigure 1a shows that 490% of mice (groups-2 and -4)conditioned with CY (300mg/kg) and BU (80 or100mg/kg) died shortly after transplantation (4–5 days).Meanwhile, all animals in group-3 conditioned withBU–CY (100 and 200mg/kg, respectively) exhibited severeaGVHD and 80% of these mice died within 10 days afterBMT (Figure 1a). Syngeneic transplanted mice receivingidentical conditioning (group-6) survived until þ 120 days(Figure 1a).

Mice conditioned with BU–CY (80 and 200mg/kg) andallogeneic transplanted (group-1) exhibited features ofaGVHD including hunched posture (Figure 1b), hair loss(Figures 1c and d), ruffled fur (Figure 1d) and mostimportantly weight loss (Figure 1e), which led to 100%death within 60 days (median survival¼ 11 days). Nomortality (Figure 1a) or signs of GVHD (Figures 1b–e)were observed in syngeneic transplanted mice. Over 60% ofnon-transplanted mice, treated with the same doses ofBU–CY died within 21 days. During the following datagroup-1 will be allogeneic and group-5 will be syngeneicwhereas group-9 is the control group.

Mean body weightWeight loss is a good indicator of GVHD in mice.22 Allconditioned mice lost weight after chemotherapy andreached the nadir at dayþ 3 and þ 7 in the syngeneic andallogeneic setting, respectively (Figure 1e). Syngeneictransplanted mice started to regain weight from dayþ 3and reached a control level within 18 days. Mean body


809


weight at day þ 5 was 16.5±1.2 and 18.4±1.3 g in group 1and group 5, respectively (Po0.001).

The allogeneic group continued to lose weight until dayþ 7 and the differences between allogeneic and syngeneicgroups remained significant until the end of the study(Po0.001) (Figure 1e). A hunched posture (Figure 1b) wasseen at day þ 7 in the allogeneic group and was correlatedwith the nadir of weight loss (Po0.001).

Histopathologic evaluation of conditioning and acuteGVHDHistopathologic effects of conditioning regimen. At day 0,a histopathological survey of tibial BM showed extensivecellular depletion with few remaining megakaryocytes. Themarrow space primarily consisted of blood filled dilatedvessels and white adipose cells. The thymus cortex wasessentially devoid of thymocytes, whereas the thymic

AlloSynCont

01214

16

18

20

22

24

-8 -6 -4 -2 0 2 4 6 8 10 14 21Days before and after BMT

Bo

dy

wei

gh

t (g

)

Su

rviv

al (

freq

uen

cy)

1.0

0.8

0.6

0.4

0.2

0.0

GroupsGroup 1Group 2Group 3Group 4Group 5Group 6Group 7Group 8Group 9

Times (days)0 20 40 60 80 100 120

Figure 1 (a) Survival analysis of different treatment groups. (’) Group 1: Female Balb/C conditioned with BU–CY (80–200mg/kg) and transplanted withmale C57BL/6; (�) group 2: female Balb/C conditioned with BU–CY (80–300mg/kg) and transplanted with male C57BL/6; (m) group 3: female Balb/Cconditioned with BU–CY (100–200mg/kg) and transplanted with male C57BL/6; ( ) group 4: female Balb/C conditioned with BU–CY (100–300mg/kg)and transplanted with male C57BL/6; (&) group 5: female Balb/C conditioned with BU–CY (80–200mg/kg) and transplanted with female Balb/C; (J)group 6: female Balb/C conditioned with BU–CY (100–200mg/kg) and transplanted with female Balb/C; (W) group 7: female C57BL/6 conditioned withBU–CY (80–200mg/kg) and transplanted with male C57BL/6; ( ) group 8: female Balb/C conditioned with BU–CY (80–200mg/kg) and not transplanted;and ( ) group 9: female C57BL/6 injected with vehicle (DMSO) only and not transplanted. All transplanted mice received 20� 106 bone marrow cells and30� 106 spleen cells from donor origin. (b–e) Manifestation of acute GVHD. Female BALB/c mice were conditioned with BU–CY (80 and 200mg/kg,respectively). At day 0 mice received 20� 106 BM and 30� 106 spleen cells from male C57BL/6 or female BALB/c (allogeneic and syngeneic setting,respectively). Starting from day þ 7 the following GVHD signs were observed: (b) hunch posture, (c) hair loss, (d) skin lesions, ulceration and denuding werethe last manifestation 2–3 days before death and (e) weight loss started after conditioning and BMT in allogeneic and syngeneic groups. (E): Allogeneicgroup, ( ): syngeneic group and (J): control mice. All values are mean±s.e. for 3–7 animals in each group per time point. Differences were analyzedstatistically employing U-test, compared between groups (*Po0.05).


810


medulla was populated by thymocytes. In the spleen, thered pulp showed diminished extramedullarly hemato-poiesis. The white pulp was populated with lymphocytes,although cellularity appeared to be decreased. The dorsalskin did not exhibit any major changes, the hair folliclesbeing in the resting telogen stage. No extensive changeswere found in the liver. Disturbance of the goblet cellpopulation was seen in colonic mucosa.

Histopathologic effects of GVHDSkin. At dayþ 7 allogeneic mice displayed fairly extensiveskin changes (Figure 2). The epidermis of the dorsal skinshowed frequent apoptotic figures (Figure 3). Small patchesof dorsal skin interfollicular epidermis appeared to havelost the basal living cell layers, only the cornified partspersisting. No hypodermal fat layer could be discerned(Figure 2). In contrast, syngeneic mice, displayed anessentially intact epidermis, whereas hypodermal fat wasseen to a varying extent (Figure 2). By day þ 14 syngeneicmice showed a normal morphology, the different stratifiedepithelia characterized by well ordered layering andmaturation. In allogeneic mice, the dorsal epidermisshowed well developed interfollicular epidermal hyperpla-sia, whereas apoptotic figures were mainly localized to hairfollicles. At dayþ 21, syngeneic mice were essentiallynormalized (Figure 2), whereas allogeneic mice, showedfeatures of extensive aGVHD (Figure 2). The interfollicularepidermal hyperplasia was extensive and combined withhyperkeratosis, in areas with immature features likehyperparakeratosis. Extensive, but localized ulcerative

changes with formation of corneal pustules were frequent.The basal layers showed intercellular edema. Apoptosis wasprevalent in the epidermis and dead cells with apoptoticfeatures seen in the hair follicles (Figure 3). Dense infiltratesof neutrophilic granulocytes were often extending intohypodermal fat.

Large intestine. By dayþ 7, allogeneic mice showed clearevidence of colitis. Crypt structure was changed andincreased presence of inflammatory cells noted in thelamina propria (Figure 2). At dayþ 21, crypts were severelydisturbed, showing hyperplasia and hyperchromasi, withnumerous cells exhibiting apoptotic features (Figure 3) andfrequent mitotic figures. In contrast, syngeneic miceexhibited near normal structure and few apoptotic cells(Figures 2 and 3).

Liver. Hepatocytes were generally smaller in the allo-geneic group. Although extramedullar hematopoiesis wasseen early, by 3 weeks liver structure was normalized in thesyngeneic, the allogeneic transplanted mice primarilyshowed increased inflammatory infiltrates in the portaltriad (Figure 2).

Immunohistochemistry evaluation of affected organ becauseof aGVHDBoth CD3þ T cells and the activated form of Caspase 3(aCPP3þ ) were frequent in liver, skin and colon in theallogeneic setting, only few were detected in syngeneic mice

Skin

Day +7 allo Day +7 syn Day +21 synDay +21 allo

Largeintestine

Liver

Figure 2 Histopathological changes in skin, liver and large intestine. Female BALB/c mice were treated with 80 and 200mg/kg of BU and CY and at day 0were transplanted with 20� 106 BM plus 30� 106 splenocyte from male C57BL/6 or female BALB/c (allogeneic and syngeneic settings, respectively). Skin,large intestine and liver were evaluated in both allogeneic and syngeneic transplanted mice at day þ 7 (beginning of GVHD) and þ 21 (established phase ofGVHD) after BMT. Apparent histological differences (leukocyte infiltration and architecture changes) between allogeneic and syngeneic transplanted miceat day þ 7 are owing to initiation of GVHD in allogeneic setting. Histological devastation progressed by time and target tissue developed full characteristicof GVHD manifestations.


811


(Figure 3). This clearly demonstrates development offeatures compatible with aGVHD.

The effects of conditioning and GVHD on myeloid andlymphoid systemAs shown in Figures 4a and b, treatment with BU–CYinduced intensive decreases in the total number of BMCand splenocytes by 95 and 88%, respectively on day 0 whencompared with controls. BM and spleen recovered rapidlyin the syngeneic recipients to control levels by day þ 14(Figures 4a and b). However, in the allogeneic recipients,the recovery of BM and spleen was slow and at day 21reached 65 and 25%, respectively when compared withcontrol (Figures 4a and b).

Treatment with BU–CY, decreased the frequency ofCD19þ cells in the BM (Figure 4c). The frequency ofCD19þ cells returned to normal in the BM at dayþ 14 inthe syngeneic group, whereas in allogeneic recipients, cellsdid not increase and gradually disappeared. The same wasobserved in the spleen (data not shown).

Conditioned animals showed a significant increase in thefrequency of CD3þ cells in the BM that was normalizedrapidly (dayþ 7) in the syngeneic recipients whereas in theallogeneic recipients, the frequency of CD3þ cells washigher at all time points when compared with control(Figure 4d). In addition 490% of the infiltrated T cells inthe allogeneic setting were of donor origin (data notshown). In the spleen, both syngeneic and allogeneic

recipients exhibited transient increase at day 0 followedby a significant decrease in the percentage of CD3þ cellsuntil dayþ 14. The frequency of splenic CD3þ cellscontinued to decrease in the allogeneic, but not in thesyngeneic recipients (Figure 4e).

The frequency of CD4þ and CD8þ T cells increased atday 0 in the spleen (Figures 4f and g). Splenic CD4þ andCD8þ T cells decreased in the syngeneic recipients untildayþ 14 and stabilized by dayþ 21. In the allogeneicrecipients, a sharp decrease in the frequency of splenicCD4þ T cells maintained until dayþ 21. In contrast, thefrequency of splenic CD8þ T cells increased rapidly byday þ 7, followed by a continuous decrease until dayþ 21(Figures 4f and g). Furthermore, the ratio of splenicCD4þ /CD8þ T cells was significantly decreased in bothsyngeneic and allogeneic recipients by dayþ 7 (Figure 4h).This decrement was greater in the allogeneic group(Po0.05). However, this ratio normalized in the syngeneicby time and continued to decrease in the allogeneicrecipients.

Evaluation of engraftment and chimerism in differenttissues of allogeneic recipientsAs shown in Figure 5a, full chimerism (98%) in the BM wasobserved at dayþ 7, which was sustained until dayþ 21.The spleen exhibited 80% chimerism at dayþ 7, whichincreased by time and reached to 95% at dayþ 21(Figure 5a). The pattern of chimerism in the blood wassimilar to that observed in the BM (data not shown).

CD3 allo CD3 syn Casp3 allo Casp3 syn

Skin

Largeintestine

Liver

Figure 3 T-cell infiltration and apoptosis in skin, large intestine and liver. Female BALB/c mice were treated with 80 and 200mg/kg of BU and CY,respectively and were transplanted at day 0 with 20� 106 BM plus 30� 106 spleen cells from male C57BL/6 or female BALB/c (allogeneic and syngeneicsetting, respectively). The most common target organs for GVHD for example, skin, large intestine and liver were examined 21 days after transplantation.The organs were stained either for CD3 (left two columns) or for Caspase 3 (right two columns). Although T cells are spread in almost all areas of targetorgans in an allogeneic setting, syngeneic transplanted mice show normal distribution of T cells. Lymphocyte infiltration in the portal area of the liver inallogeneic transplanted mice is higher when compared with that seen in a syngeneic group. Caspase 3, an indicator of apoptosis, is more frequent in targettissues after allogeneic transplantation when compared with that seen after syngeneic transplantation.


812


-7 0 7 14 21 280

10

20

30

40

50 AlloSynCont

AlloSynCont

AlloSynCont

AlloSynCont

AlloSynCont

AlloSynCont

AlloSynCont

AlloSynCont

Days before and after BMT-7 0 7 14 21 28

Days before and after BMT

BM

cel

lula

rity

×10

6

0

50

100

150

200

250

300

Sp

leen

cel

lula

rity

×106

CD19 in BM

-7 0 7 14 21 280

10

20

30

40

50

Day

Freq

uen

cy (

%)

CD3 in BM

-7 0 7 14 21 280

2

4

6

8

10


Freq

uen

cy (

%)

CD3 in Spleen

-7 0 7 14 21 280

10

20

30

40

50


-7 0 7 14 21 28Days before and after BMT

-7 0 7 14 21 28Days before and after BMT

Freq

uen

cy (

%)

0

10

20

30

40

50

Freq

uen

cy (

%)

CD4 in spleen

CD8 in spleen

0

5

10

15

20

Freq

uen

cy (

%)

CD4 /CD8 ratio

-7 0 7 14 21 280

1

2

3

4

5

6


Rat

io

Figure 4 BM and spleen cellularity and lymphoid subpopulations after BMT. Female BALB/c mice were conditioned with 80mg/kg of BU followed by200mg/kg of CY. Treated mice received 20� 106 BM plus 30� 106 spleen cells from male C57BL/6 (allogeneic setting) or female BALB/c (syngeneic setting).A group of female BALB/c without conditioning and BMT was considered as a control group. BM (a) and spleen (b) cellularity. B-lymphocytes (CD19)frequency in BM (c). T-lymphocyte (CD3) population in both BM (d) and spleen (e). Frequency of T-helper (CD4) and cytotoxic (CD8) cells (f and g). Sevendays after BMT CD4/CD8 ratio (h) in the spleen of allogeneic transplanted mice decreased more than the syngeneic group (Po0.05). Moreover the rationormalized in syngeneic transplanted mice whereas decrement continued in allogeneic group until the end of the experiment. (E): allogeneic group,( ): syngeneic group and (J): control mice. All values are mean±s.e. for 3–7 animals in each group per time point. Differences were analyzed statisticallyemploying U-test, compared between groups (*Po0.05).


813


Furthermore, at dayþ 7, chimerism of CD11bþ andCD3þ cell populations in the spleen were 95 and 48%,respectively (Figure 5b). Chimerism increased to reach98 and 89%, respectively at dayþ 21. Although in theallogeneic recipients, reconstitution of B-cells was pro-

foundly depressed (Figure 4c); however, 490% were ofdonor origin in all tested intervals (not shown).

FISH analysis of the BM and spleen of the syngeneicgroup (using Y-chromosome) showed 81 and 90% donorcells, respectively at dayþ 7. By day þ 21, the chimerism in

Spleen and BM chimerism

7 14 210

20

40

60

80

100

Spleen

BM

Days after BMT

7 14 21Days after BMT



Do

no

r ch

imer

ism

(%

)

0

20

40

60

80

100

Do

no

r ch

imer

ism

(%

)

CD3, CD11b Chimerism in spleen

CD3

CD11b

CD3 T-cells in skin

0

1

2

3

4

5AlloSyn

Freq

uen

cy (

%)

Donor cells in recipient skin

Total Donor cells

Donor CD3 cell

0

10

2050

75

100

Do

no

r ch

imer

ism

(%

)

H2Kb

CD

3

Sid

e sc

atte

r

250

200

150

100

50

0

CD

3 P

erC

P-C

y5.5

100 101 102 103 104 100 101 102 103 104

H-2Kb PEH-2Kb PE

100

101

102

103

104

R1

R1

90.6%

9.4%

6.15%

Figure 5 BM and spleen chimerism and skin infiltration by donor T cells. Female BALB/c mice were conditioned with 80 and 200mg/kg of BU and CY,respectively and were transplanted at day 0 by 20� 106 BM plus 30� 106 SP cells from male C57BL/6 and female BALB/c (allogeneic and syngeneic setting,respectively). (a) At appointed time points, 5–7 transplanted mice were killed and total chimerism was evaluated in the spleen and BM. BM of recipientshowed full donor chimerism (495%) at day þ 7. Chimerism in spleen was about 80% at day þ 7 and reached to495% at day 21 after allogeneic BMT. (b)Seven days after BMT chimerism of T-lymphocytes (CD3þ ) and myeloid lineage (CD11bþ ) in the spleen of recipient mice was 50 and 95% respectivelyand increased to 90% in CD3þ population and sustained the same range in myeloid population during next 3 weeks after BMT. (c and d) FACS analysis ofskin showed that T cells frequency (c) in the skin of allogeneic transplanted mice was higher when compared with that observed in syngeneic group at allevaluated time points. Although more than 90% of infiltrated T cells were of donor origin (d). (e and f) Twenty-one days after BMT, intestinal single cellsanalysis using FACS showed that donor cells (H-2Kb) are present in the host small intestine (e). Likewise 490% of infiltrating donor cells were T cells (f).(E): Allogeneic group, ( ): syngeneic group and (J): control mice. All values are mean±s.e. for 3–7 animals in each group per time point. Differences wereanalyzed statistically employing U-test, compared between groups (*Po0.05).


814


the BM and spleen reached 96 and 88%, respectively andsustained (498%) in both organs to day þ 120.

Profound infiltration of immune cells in the skin andintestine was observed (Figures 2 and 3). Characterizationof infiltrating cells in the skin (Figure 3) showed that inparallel with the progression of GVHD, the frequency ofCD3þ cells increased in allogeneic, but not syngeneicrecipients (Figure 5c). Moreover, analysis of the infiltratingcells demonstrated that the frequency of total and CD3þcells in the skin increased by time and 490% of theinfiltrating CD3þ cells were of donor origin at all testedtime points (Figure 5d). In addition, analysis of recipientintestinal cells at dayþ 21 showed that over 6% of wholecell populations were of donor origin (Figure 5f). Moreover490% of the infiltrating donor cells were T-lymphocytes(Figure 5f), which might explain the pathophysiology of gutinjury due to GVHD. These results are in line withhistopathological finding of skin and intestine.

Discussion

In the present investigation, we introduce for the first time amouse model of aGVHD based on chemotherapy con-ditioning. During the last three decades, numerous modelsof aGVHD have been used to explore mechanisms under-lying the development of GVHD.23,24 Almost all currentmodels use TBI alone as conditioning regimen beforeHSCT.25–27 However, several clinical studies have reportedthe crucial role of conditioning in the development andseverity of aGVHD.28,29 On the other hand 450% ofpatients are conditioned with chemotherapy-based regi-mens. Unfortunately, scant knowledge about chemother-apy-based conditioning in experimental models is available.

Previous studies have shown that BU treatment with100–150mg/kg is sufficient to induce myeloablation and/orfull donor chimerism.30,31 Thus the lack of donor allo-reactivity against host residual cells and/or the high dose ofBU may explain the lack of need of CY. In our previousstudies,32,33 we have shown that similar myeloablativeeffects of BU were obtained after the administration of BUas i.v. or i.p., therefore, BU was administered i.p. in thepresent investigation. In our dose finding studies, a dose of4� 20mg/kg followed by 2� 100mg/kg CY was necessaryto reach full donor chimerism. Moreover, increasing thedose of BU or CY or both have resulted in a rapidemergence of aGVHD implying that chemotherapy mayhave a pivotal role in the severity of aGVHD.

Clinical manifestations and histological changes oc-curred between dayþ 7 and dayþ 14, resembled aGVHDdescribed in irradiated animals in allogeneic transplanta-tion.34,35 The weight loss is the most well documentedmanifestation of GVHD in mice.18,22,36 In irradiation-basedaGVHD models, it has been shown that the weight lossoccurs in a bi-phasic manner, that is, a severe phaseculminates on dayþ 7 followed by an improvement in theweight until dayþ 12 and then a second phase of weightloss commences on dayþ 16.37,38 Interestingly, in thepresent model, we observed a bi-phasic pattern in theweight loss. However, the first phase started by a modestweight loss at dayþ 3, followed by a severe decrement

phase at dayþ 7. Stable weight at days þ 10 to þ 14 wasfollowed by a second phase of weight loss at dayþ 21. Thedelay in the occurrence of the second phase of weight loss isprobably because of differences in myeloablative andimmunosuppressive effects after BU–CY compared withthat observed after TBI. The present results add moreevidence that different pre-transplant conditioning leads tovarious organ damage and different toxicity profiles.

Skin, gastrointestinal tract, liver and lymphoid organsare the primary target organs in acute GVHD.27,39,40 Mostof the pathological alterations including infiltration ofinflammatory cells into the skin, liver and large intestineand the presence of high numbers of apoptotic/necroticcells in all tested organs were similar to those observed inirradiation-based aGVHD mouse models.40–42 Our resultsconfirm that the present model fulfils the criteria ofaGVHD including histopathological changes.

The mechanism underlying the development of aGVHDin certain organs is not fully understood. However, in a3-phase model proposed for the pathogenesis ofaGVHD,43–45 including phase 1, where the conditioningregimen leads to the initial damage and activation ofdendritic cells of the host tissues which provides a conditionfor the migration and activation of donor T cells to thetarget organ tissue in phase-2.43–45 The importance ofdonor T cells in target organ pathology has been shownearlier.46–48 In a mouse model of GVHD based onirradiation, Baker et al. and Nikolic et al.47,48 have reportedskin and intestine damage in the allogeneic setting and theyshowed the presence of T cells in the skin lesion and GI;however, they did not confirm the origin of these cells. Inthe present investigation both immunohistochemistry andFACS results have shown that donor T cells were found inthe skin and intestine. FACS analysis showed that thenumber of infiltrating cells is correlated with severity oflesions.

Finally, phase 3 (the effector phase) involves a complexcascade of multiple cellular and inflammatory effects thatcause severe changes in target organs.43–45 Our resultsshowed pathological alterations in skin and large intestine,but not the liver at day 0 which probably is because theliver is highly specialized to detoxify xenobiotics, and thatthe threshold of BU–CY induced toxicity in the liver ishigher than that for skin, large intestine and lymphoidorgans. Several studies have shown that donor T cells areactivated by recipients/donor antigen-presenting cells in thesecondary lymphoid organs such as the spleen, lymphnodes and Peyer’s patches.49,50 In this study, we have alsoobserved that the frequency of donor T cells (particularlyCD8þ ) is increased dramatically in the spleen at thebeginning of aGVHD. Subsequently, by the progressionand spreading of GVHD to target organs, the level ofdonor T cells decreased in the spleen, but increased in thetarget organs. This data is in agreement with that observedby Panoskaltsis et al. and Beilhack et al.34,49 where theauthors tracked the trafficking of donor T cells in the hostand they showed the importance of secondary lymphoidorgans in initiation of GVHD.50 Moreover Muraiet al.51and others have shown the higher priority of Peyer’spatch rather than the spleen in stimulation and allo-reactivation of donor lymphocyte.52 Our data are contra-


815


dictory to that reported by Clouthier et al.53 who developedaGVHD in splenectomized mice. This is most likelybecause other secondary lymphoid organs may compensatefor the spleen.

Several studies54,55 have shown that in MHC mismatchtransplantation, both CD4þ and CD8þ alloreactive Tcells can induce aGVHD. In the present study, we found anintensive expansion of CD8þ T cells in the spleen ofrecipient mice at dayþ 7. Moreover, a continuous decreaseof the ratio CD4þ /CD8þ splenic T cells was observed inthese animals. These findings and the presence of highlevels of apoptosis in the skin, liver and large intestinesuggest that CD8þ T cells have an important function inthe effector phase of aGVHD.

BM56 and secondary lymphoid organs, namely spleen48,57

are the first targets for GVHD. BM suppression inallogeneic BMT was reported to be caused by cytotoxic Tcells.58 In our study, the cellularity of BM and spleen weresignificantly (Po0.05) lower in allogeneic compared withsyngeneic group. Van Dijken et al.59 suggested that GVHDdoes not affect the BM cellularity of the recipient.However, the authors showed that GVHD decreases thehematopoietic reserve and causes damage to the donorstem cell compartment. In our study the cellularity wasaltered by GVHD which might be because of the majorhistocompatibility mismatch used in our model thatinduced severe GVHD compared with the published studywhere minor histocompatability was used.

An interesting finding in the present model was thatduring the development of aGVHD, the frequency of B-cells was considerably decreased in both BM and spleen.These observations imply that during the development ofaGVHD, a profound impairment occurs in the B-lympho-poieses, which leads to an acquired B-cell deficiency.Although, to our knowledge, this phenomenon has notbeen reported in GVHD models, clinical studies haveshown that patients with GVHD have a decreased B-lymphopoiesis.60 The underlying mechanisms for theimpaired B-lymphopoiesis are not known. However, ithas been suggested that this phenomenon occurs as a resultof either the production of cytokines from the activatedalloreactive T cells or the destruction of stromal cells, whichsupport B-lymphopoiesis.61

To conclude, the present model has an importantfunction for future studies regarding GVHD immunobio-logy, evaluation-conditioning regimen toxicity and impor-tantly an animal model for the variety of patients under-going chemotherapy-based conditioning.

Acknowledgements

We thank The Swedish Cancer foundation ‘Cancer Fonden’ forthe support provided for this investigation.

References

1 Thomas ED. Stem cell transplantation: past, present andfuture. Stem Cells 1994; 12: 539–544.

2 Storb R. Allogeneic hematopoietic stem cell transplantation—yesterday, today, and tomorrow. Exp Hematol 2003; 31: 1–10.

3 Ferrara JL, Deeg HJ. Graft-versus-host disease. N Engl J Med1991; 324: 667–674.

4 Mielcarek M, Martin PJ, Leisenring W, Flowers ME, MaloneyDG, Sandmaier BM et al. Graft-versus-host disease afternonmyeloablative versus conventional hematopoietic stem celltransplantation. Blood 2003; 102: 756–762.

5 Tabbara IA. Allogeneic bone marrow transplantation: acuteand late complications. Anticancer Res 1996; 16: 1019–1026.

6 Blazar BR, Taylor PA, Snover DC, Bluestone JA, Vallera DA.Nonmitogenic anti-CD3F(ab’)2 fragments inhibit lethalmurine graft-versus-host disease induced across the majorhistocompatibility barrier. J Immunol 1993; 150: 265–277.

7 Terwey TH, Kim TD, Kochman AA, Hubbard VM, Lu S,Zakrzewski JL et al. CCR2 is required for CD8-induced graft-versus-host disease. Blood 2005; 106: 3322–3330.

8 Ferrara JL, Cooke KR, Teshima T. The pathophysiologyof acute graft-versus-host disease. Int J Hematol 2003; 78:181–187.

9 Deeg HJ. Cytokines in graft-versus-host disease and the graft-versus-leukemia reaction. Int J Hematol 2001; 74: 26–32.

10 Clift RA, Buckner CD, Appelbaum FR, Bearman SI, PetersenFB, Fisher LD et al. Allogeneic marrow transplantation inpatients with acute myeloid leukemia in first remission: arandomized trial of two irradiation regimens. Blood 1990; 76:1867–1871.

11 Santos GW. Preparative regimens: chemotherapy versuschemoradiotherapy. A historical perspective. Ann N Y AcadSci 1995; 770: 1–7.

12 Wiebe VJ, Smith BR, DeGregorio MW, Rappeport JM.Pharmacology of agents used in bone marrow transplantconditioning regimens. Crit Rev Oncol Hematol 1992; 13:241–270.

13 Vriesendorp HM. Aims of conditioning. Exp Hematol 2003;31: 844–854.

14 Thomas ED, Buckner CD, Banaji M, Clift RA, Fefer A,Flournoy N et al. One hundred patients with acute leukemiatreated by chemotherapy, total body irradiation, and allo-geneic marrow transplantation. Blood 1977; 49: 511–533.

15 Santos GW, Tutschka PJ, Brookmeyer R, Saral R, BeschornerWE, Bias WB et al. Marrow transplantation for acutenonlymphocytic leukemia after treatment with busulfan andcyclophosphamide. N Engl J Med 1983; 309: 1347–1353.

16 Ferry C, Socie G. Busulfan-cyclophosphamide versus totalbody irradiation-cyclophosphamide as preparative regimenbefore allogeneic hematopoietic stem cell transplantation foracute myeloid leukemia: what have we learned? Exp Hematol2003; 31: 1182–1186.

17 Ringden O, Ruutu T, Remberger M, Nikoskelainen J, Volin L,Vindelov L et al. A randomized trial comparing busulfan withtotal body irradiation as conditioning in allogeneic marrowtransplant recipients with leukemia: a report from theNordic Bone Marrow Transplantation Group. Blood 1994;83: 2723–2730.

18 Cooke KR, Kobzik L, Martin TR, Brewer J, Delmonte Jr J,Crawford JM et al. An experimental model of idiopathicpneumonia syndrome after bone marrow transplantation: I.The roles of minor H antigens and endotoxin. Blood 1996; 88:3230–3239.

19 Perl A. Pathogenesis and spectrum of autoimmunity. MethodsMol Med 2004; 102: 1–8.

20 Lefrancois L, Lycke N. Isolation of mouse small intestinalintraepithelial lymphocytes, Peyer’s patch, and laminapropria cells. In: Coligan JE, Kruisbeek AM, Margulies DH,Shevach EM, Storber W (eds). Current Protocols in Immuno-logy, Vol. 1. John Wiley & Sons, Inc.: Hoboken, NJ,USA, 1996. Unite no 3.19 pages (3.19.1–3.19.16) (ISBN 0-471-52276-7).


816


21 Nilsson C, Forsman J, Hassan Z, Abedi-Valugerdi M,O’Connor C, Concha H et al. Effect of altering administrationorder of busulphan and cyclophosphamide on the myelo-ablative and immunosuppressive properties of the conditioningregimen in mice. Exp Hematol 2005; 33: 380–387.

22 Fontaine P, Perreault C. Diagnosis of graft-versus-host diseasein mice transplanted across minor histocompatibility barriers.Transplantation 1990; 49: 1177–1179.

23 Blazar BR, Murphy WJ. Bone marrow transplantation andapproaches to avoid graft-versus-host disease (GVHD). PhilosTrans R Soc Lond B Biol Sci 2005; 360: 1747–1767.

24 Korngold R, Sprent J. Graft-versus-host disease in experi-mental allogeneic bone marrow transplantation. Proc Soc ExpBiol Med 1991; 197: 12–18.

25 Dutt S, Ermann J, Tseng D, Liu YP, George TI, Fathman CGet al. L-selectin and beta7 integrin on donor CD4 T cells arerequired for the early migration to host mesenteric lymphnodes and acute colitis of graft-versus-host disease. Blood2005; 106: 4009–4015.

26 Rosset MB, Tieng V, Charron D, Toubert A. Differences inMHC-class I presented minor histocompatibility antigensextracted from normal and graft-versus-host disease (GVHD)mice. Clin Exp Immunol 2003; 132: 46–52.

27 Ferrara J, Guillen FJ, Sleckman B, Burakoff SJ, Murphy GF.Cutaneous acute graft-versus-host disease to minor histocompat-ibility antigens in a murine model: histologic analysis andcorrelation to clinical disease. J Invest Dermatol 1986; 86: 371–375.

28 Hill GR, Crawford JM, Cooke KR, Brinson YS, Pan L,Ferrara JL. Total body irradiation and acute graft-versus-hostdisease: the role of gastrointestinal damage and inflammatorycytokines. Blood 1997; 90: 3204–3213.

29 Nash RA, Pepe MS, Storb R, Longton G, Pettinger M,Anasetti C et al. Acute graft-versus-host disease: analysis ofrisk factors after allogeneic marrow transplantation andprophylaxis with cyclosporine and methotrexate. Blood 1992;80: 1838–1845.

30 Ploemacher RE, Johnson KW, Rombouts EJ, Etienne K,Westerhof GR, Baumgart J et al. Addition of treosulfan to anonmyeloablative conditioning regimen results in enhancedchimerism and immunologic tolerance in an experimentalallogeneic bone marrow transplant model. Biol Blood MarrowTransplant 2004; 10: 236–245.

31 Jopling C, Rosendaal M. A cautionary tale: how to deletemouse haemopoietic stem cells with busulphan. Br J Haematol2001; 113: 970–974.

32 Hassan Z, Hellstrom-Lindberg E, Alsadi S, Edgren M,Hagglund H, Hassan M. The effect of modulation ofglutathione cellular content on busulphan-induced cytotoxicityon hematopoietic cells in vitro and in vivo. Bone MarrowTransplant 2002; 30: 141–147.

33 Hassan Z, Nilsson C, Hassan M. Liposomal busulphan:bioavailability and effect on bone marrow in mice. BoneMarrow Transplant 1998; 22: 913–918.

34 Panoskaltsis-Mortari A, Price A, Hermanson JR, Taras E,Lees C, Serody JS et al. In vivo imaging of graft-versus-host-disease in mice. Blood 2004; 103: 3590–3598.

35 Hamilton BL, Parkman R. Acute and chronic graft-versus-host disease induced by minor histocompatibility antigens inmice. Transplantation 1983; 36: 150–155.

36 Korngold B, Sprent J. Lethal graft-versus-host disease afterbone marrow transplantation across minor histocompatibilitybarriers in mice. Prevention by removing mature T cells frommarrow. J Exp Med 1978; 148: 1687–1698.

37 Fukui J, Inaba M, Ueda Y, Miyake T, Hosaka N, Kwon AHet al. Prevention of graft-versus-host disease by intra-bone marrow injection of donor T cells. Stem Cells 2007; 25:1595–1601.

38 van Leeuwen L, Guiffre A, Atkinson K, Rainer SP, SewellWA. A two-phase pathogenesis of graft-versus-host disease inmice. Bone Marrow Transplant 2002; 29: 151–158.

39 Heldal D, Brinch L, Evensen SA, Tjonnfjord GE, Aamodt G,Elgjo K et al. Skin biopsies for early diagnosis and prognosisof graft-versus-host disease in recipients of allogeneic stem cellsfrom blood or bone marrow. Bone Marrow Transplant 2004;34: 345–350.

40 Goker H, Haznedaroglu IC, Chao NJ. Acute graft-vs-hostdisease: pathobiology and management. Exp Hematol 2001;29: 259–277.

41 Yoo YH, Gilliam AC, Whitaker-Menezes D, Korngold R,Murphy GF. Experimental induction and ultrastructuralcharacterization of apoptosis in murine acute cutaneousgraft-versus-host disease. Arch Dermatol Res 1997; 289:389–398.

42 Piguet PF, Grau GE, Allet B, Vassalli P. Tumor necrosisfactor/cachectin is an effector of skin and gut lesions of theacute phase of graft-vs.-host disease. J Exp Med 1987; 166:1280–1289.

43 Sun Y, Tawara I, Toubai T, Reddy P. Pathophysiology ofacute graft-versus-host disease: recent advances. Transl Res2007; 150: 197–214.

44 Hofmeister CC, Quinn A, Cooke KR, Stiff P, Nickoloff B,Ferrara JL. Graft-versus-host disease of the skin: life anddeath on the epidermal edge. Biol Blood Marrow Transplant2004; 10: 366–372.

45 Hill GR, Ferrara JL. The primacy of the gastrointestinal tractas a target organ of acute graft-versus-host disease: rationalefor the use of cytokine shields in allogeneic bone marrowtransplantation. Blood 2000; 95: 2754–2759.

46 Dutt S, Tseng D, Ermann J, George TI, Liu YP, Davis CRet al. Naive and memory T cells induce different types of graft-versus-host disease. J Immunol 2007; 179: 6547–6554.

47 Nikolic B, Lee S, Bronson RT, Grusby MJ, Sykes M. Th1 andTh2 mediate acute graft-versus-host disease, each with distinctend-organ targets. J Clin Invest 2000; 105: 1289–1298.

48 Baker MB, Altman NH, Podack ER, Levy RB. The role ofcell-mediated cytotoxicity in acute GVHD after MHC-matched allogeneic bone marrow transplantation in mice.J Exp Med 1996; 183: 2645–2656.

49 Beilhack A, Schulz S, Baker J, Beilhack GF, Wieland CB,Herman EI et al. In vivo analyses of early events in acute graft-versus-host disease reveal sequential infiltration of T-cellsubsets. Blood 2005; 106: 1113–1122.

50 Shlomchik WD, Couzens MS, Tang CB, McNiff J, RobertME, Liu J et al. Prevention of graft versus host disease byinactivation of host antigen-presenting cells. Science 1999; 285:412–415.

51 Murai M, Yoneyama H, Ezaki T, Suematsu M, Terashima Y,Harada A et al. Peyer’s patch is the essential site in initiatingmurine acute and lethal graft-versus-host reaction. NatImmunol 2003; 4: 154–160.

52 Guy-Grand D, Vassalli P. Gut injury in mouse graft-versus-host reaction. Study of its occurrence and mechanisms. J ClinInvest 1986; 77: 1584–1595.

53 Clouthier SG, Ferrara JL, Teshima T. Graft-versus-hostdisease in the absence of the spleen after allogeneicbone marrow transplantation. Transplantation 2002; 73:1679–1681.

54 Korngold R, Sprent J. T cell subsets and graft-versus-hostdisease. Transplantation 1987; 44: 335–339.

55 Sprent J, Schaefer M, Lo D, Korngold R. Properties ofpurified T cell subsets. II. In vivo responses to class I vs class IIH-2 differences. J Exp Med 1986; 163: 998–1011.

56 Anderson KC, Weinstein HJ. Transfusion-associated graft-versus-host disease. N Engl J Med 1990; 323: 315–321.


817


57 Tschetter JR, Mozes E, Shearer GM. Progression from acuteto chronic disease in a murine parent-into-F1 model of graft-versus-host disease. J Immunol 2000; 165: 5987–5994.

58 Piguet PF. GVHR elicited by products of class I or class II lociof the MHC: analysis of the response of mouse T lymphocytesto products of class I and class II loci of the MHC incorrelation with GVHR-induced mortality, medullary aplasia,and enteropathy. J Immunol 1985; 135: 1637–1643.

59 van Dijken PJ, Wimperis J, Crawford JM, Ferrara JL. Effectof graft-versus-host disease on hematopoiesis after bonemarrow transplantation in mice. Blood 1991; 78: 2773–2779.

60 Storek J, Wells D, Dawson MA, Storer B, Maloney DG.Factors influencing B lymphopoiesis after allogeneic hemato-poietic cell transplantation. Blood 2001; 98: 489–491.

61 LeBien TW. Fates of human B-cell precursors. Blood 2000; 96:9–23.


818


GVHD after chemotherapy conditioning in allogeneic transplanted mice

Documents

Transcript of GVHD after chemotherapy conditioning in allogeneic transplanted mice