Kidney International, Vol. 56 (1999), pp. 1101–1112

Peripheral donor leukocytes prolong survival of ratrenal allografts

MARINA NORIS, NADIA AZZOLLINI, MARILENA MISTER, ANGELA PEZZOTTA,GIAMPIERO PICCININI, FEDERICA CASIRAGHI, DANIELA CUGINI, NORBERTO PERICO,SILVIA ORISIO, and GIUSEPPE REMUZZI

Department of Immunology and Clinics of Organ Transplantation, Ospedali Riuniti Bergamo—Mario Negri Institute forPharmacological Research, Bergamo, Italy

Peripheral donor leukocytes prolong survival of rat renal allo- The ultimate goal in allotransplantation is modulationgrafts. of the immune response to produce tolerance to the graft

Background. The development of strategies to enhance the without the need of systemic immunosuppression [1]. Insurvival of transplanted organs and to potentially lower or evenrecent clinical studies, concomitant donor bone marrowdiscontinue immunosuppressive therapy would represent a sig-

nificant advancement in post-transplant patient care. infusion at the time of cadaveric transplantation has beenMethods. We studied the effect of pretransplant infusion of attempted in humans as a means to improve the survival

donor leukocytes alone or in combination with a short course of vascularized organs [2–5] via enhancing microchimer-of cyclosporine on the long-term outcome of a rat model of

ism, that is, the natural migration of bone marrow-kidney allograft.derived donor leukocytes from the allograft to the recipi-Results. A single intravenous infusion of donor peripheral

blood leukocytes (100 3 106 cells) from Brown-Norway (BN) ent organs and tissues [6–8]. The strategy of infusingrats into major histocompatibility complex (MHC) incompati- donor bone marrow at the same time or shortly afterble Lewis recipients largely failed to prolong kidney allograft transplantation of whole organs is merely an augmen-viability from the same donor transplanted 60, 40, or 30 days

tation of the normal post-transplant cell migration [8, 9].after cell infusion. A short course of cyclosporine (per se, unableto prolong graft survival) was started at the same day of donor Concomitant donor bone marrow and organ transplanta-leukocyte infusion, but instead was able to prolong the survival tion enhances the amount of the so-called “passenger”of the BN kidney transplant—performed 40 days later—but leukocytes, which are cells consisting mainly of bonenot of a Wistar Furth (WF) third party, with some animals

marrow-derived dendritic cell progenitors [6–8]. Theyeven developing tolerance. A mixed lymphocyte reaction ofare capable of homing to secondary lymphoid tissueshost cells from long-term surviving rats to BN stimulator cells

was significantly reduced as compared with controls. Donor and to present antigen to T cells therein [10–13]. ThereBN DNA was detected in the peripheral blood of Lewis rats has been increased speculation that such immature den-until day 40 after BN leukocyte infusion. Microchimerism per- dritic cells present in nonlymphoid organs may constitutesisted (60 to 70 days post-transplant) in most long-term graft

tolerogenic progenitors of chimeric cells in recipient ofrecipients. Reducing the time interval between donor leukocyteorgan allografts [7–9]. Enhancing microchimerism by us-infusion and subsequent kidney transplant to 10 days still pro-

longed graft survival. Donor peripheral blood mononuclear ing bone marrow, however, predisposes individuals tocells, but not polymorphonuclear cells, in the leukocyte prepa- graft-vs.-host disease (GVHD), as has been documentedration contributed to prolong kidney allograft survival.

in recent clinical trials [4, 5], given its multilineage cellConclusions. Pretransplant donor leukocyte infusion undercomposition, including T cells that drive the adversethe appropriate conditions can tip the immune balance toward

improved graft acceptance. This result could be relevant to the reaction against the graft recipient. Moreover, with thisachievement of donor-specific tolerance of the graft with the strategy, tolerance has been achieved only in animalmaintenance of an intact response to third-party antigens. recipients conditioned with a myeloablative [14] or non-

myeloablative [15, 16] dose of whole body irradiation topermit major histocompatibility complex (MHC)–mis-Key words: microchimerism, transplantation, cyclosporine, rat MHC

class II gene, mononuclear leukocyte. matched bone marrow engraftment, which, however,precludes its use in clinical organ transplantation. On

Received for publication July 16, 1998the other hand, immunosuppressive therapy employedand in revised form February 15, 1999

Accepted for publication April 5, 1999 in clinical trials of bone marrow and organ transplanta-tion to augment chimerism does not permit any conclu- 1999 by the International Society of Nephrology

1101

Noris et al: Donor leukocytes and kidney allograft survival1102

sion on the possibility of achieving a drug-free allograft- 106 per rat) as mentioned earlier in this article. Startingon the day of cell infusion, Lewis rats also received atolerant state.

In rats, combining multiple organ grafts (two hearts daily intramuscular (i.m.) injection of 10 mg/kg/day CsA(Novartis, Basel, Switzerland) for 14 days and were sup-and two kidneys, for instance) with peritransplant infu-

sion of donor spleen leukocytes prolonged incompatible plemented with single injections on days 21 and 28. Fortydays after leukocyte infusion, these animals underwentgraft survival for more than 200 days [17], but leaves

unanswered the question of whether the observed tolero- kidney transplant from the same BN donor strain. Asimilar protocol was followed with native Lewis rats whogenic effect was actually caused by the large mass of

donor tissue or to the simultaneous load of leukocytes. were treated with CsA but did not receive BN leukocytes(group 6, N 5 6). The time schedule of administrationWith this background, our study was designed to ex-

plore the property of pretransplant infusion of donor of CsA was chosen according to previous findings ofsuccessful engraftment with this regimen but using ta-leukocytes alone or in combination with a short course

of cyclosporine (CsA), to possibly prolong graft survival crolimus instead of CsA [18]. Moreover, to assesswhether the potential tolerogenic effect of the peripheralin a rat model of single kidney allograft as an alternative

to bone marrow infusion. blood leukocytes is donor specific, two additional groupsof animals were considered. In group 7 (N 5 4), Lewisrats underwent the same protocol as discussed earlier in

METHODSthis article, but received a third-party WF kidney allo-

Animals graft 40 days after BN peripheral blood leukocyte infu-sion. In group 8 (N 5 4), Lewis rats received peripheralInbred male rats weighing 175 to 200 g were used in

all experiments (Charles River Italia S.p.a., Calco, Italy). blood leukocytes (100 3 106 per experiment) from third-party WF rats and 40 days later received a BN kidneyLewis rats (LW, RT1l) acted as recipients, and Brown-

Norway rats (BN, RT1n) or Wistar Furth (WF, RT1u) allograft. An additional control group of Lewis rats(group 9, N 5 3) did receive the BN kidney allograftacted as donors. Animal care and treatment were con-

ducted in conformity with institutional guidelines in com- alone.In all of these studies, kidney graft function was moni-pliance with national and international laws and policies

(EEC Council Directive 86/609, OJL 358, December tored by determining the 24-hour urine output in individ-ual metabolic cages every other day during the first week1987; “Guide for the Care and Use of Laboratory Ani-

mals,” U.S. National Research Council 1996). Animals of post-transplantation. Blood samples were also col-lected for measurement of the serum creatinine levelwere maintained in conventional animal facilities and

had free access to standard rat chow and tap water. (Reflotont; Boehringer Mannheim, Milan, Italy) twicea week or at sacrifice or every month post-transplant.

Experimental design Operational tolerance was determined by kidney graftsurvival considered synonymous with recipient death, orWe first evaluated the potential role of donor periph-

eral blood leukocytes to prolong graft survival by infus- sacrifice before then because of its moribund state.Peripheral blood mononuclear cells (PBMCs) froming the preparation at different time points before kidney

transplantation without any immunosuppressants. Thus, long-term surviving recipients who were previously givenBN peripheral blood leukocytes and kidney from thenaive Lewis rats were given an intravenous injection of

BN peripheral blood leukocytes (100 3 106 per rat) under same donor strain were tested for their ability to prolifer-ate in vitro to BN spleen cells in a standard one-waylight anesthesia 60 days (group 1, N 5 4), 40 days (group

2, N 5 5), and 30 days (group 3, N 5 7) before kidney mixed lymphocyte reaction (MLR) assay. As a control,the MLR response of PBMCs from Lewis rats that re-transplantation from the same BN donor strain. As a

control (group 4, N 5 3), Lewis rats received a BN ceived BN kidney allografts but were pretreated withWF peripheral blood leukocytes (third party) to BNkidney allograft alone. Graft function was monitored

thereafter. spleen cells was performed.To assess whether the short course of CsA may addWe then explored the possibility that concomitant

short-course immunosuppression at the time of donor benefit to the potential tolerogenic effect of donor leuko-cyte infusion by enhancing cell engraftment in the recipi-leukocyte infusion may afford any additional benefit to

the potential tolerogenic effect of this strategy. To ad- ents, cellular microchimerism in the peripheral bloodwas serially determined at given time points after donordress this issue, experiments were performed using an

approach similar to that recently reported by Murase et cell infusion. Lewis rats received 100 3 106 donor BNperipheral blood leukocytes with (group 10, N 5 3) oral, with combined bone marrow infusion and liver or

heart transplantation in a Lewis to BN rat combination without (group 11, N 5 3) a concomitant short courseof CsA. Blood was collected from a peripheral vein at[18]. Lewis rats (group 5, N 5 16) were given an intrave-

nous injection of BN peripheral blood leukocytes (100 3 5 minutes, 1 hour, and 1, 2, 10, 20, and 40 days after

Noris et al: Donor leukocytes and kidney allograft survival 1103

donor BN leukocyte infusion, and chimerism evaluated placed in an iced saline solution until transplant. Lewisrecipients were prepared by removal of the left kidney.as described later in this article. Moreover, the presence

of microchimerism was also determined 60 to 70 days An anastomosis was created between the recipient andthe donor renal artery, as well as the renal vein withpost-transplant in the peripheral blood of six long-term

kidney graft recipients previously infused with BN donor end-to-end anastomosis. The vascular clamps were re-leased as soon as the vascular anastomoses were com-leukocytes.

We also examined whether or not repeated adminis- pleted, with an ischemia time of 20 to 30 minutes. Donorand recipient ureters were attached end to end. Thetration of donor leukocytes to the naive Lewis rats in

order to enhance the antigen loading further may sensi- native right kidney was then removed, and the rats wereallowed to recover in their cages.tize the recipients and differently affect kidney graft sur-

vival. To this purpose, Lewis rats received two intrave-Isolation and preparation of donor leukocytesnous infusions of 100 3 106 peripheral blood leukocytes

from BN rats at day 0 and seven days later (day 7) Leukocytes were isolated from heparinized venousblood collected by cardiac puncture of BN rats. Bloodin the study protocol (group 12, N 5 8). The time of

administration of CsA was the same as for the single was mixed with Emagelt (Hoechst Marion Roussel, Mi-lan, Italy), and erythrocytes were allowed to sediment.infusion of donor cells. Forty days after the first cell

infusion, the animals underwent BN kidney transplant, The leukocyte-enriched supernatant obtained after 40minutes of incubation at 48C was then removed andand graft survival was monitored thereafter, as discussed

earlier in this article. centrifuged for 8 minutes at 48C. The remaining erythro-cytes in the cell pellet were removed by ammonium chlo-Next, we attempted to reduce the time interval be-

tween donor cell infusion and transplantation. Thus, ride lysis at 48C and subsequent centrifugation.Peripheral blood mononuclear cells were isolatedLewis rats (group 13, N 5 5) were given 100 3 106 BN

peripheral blood leukocytes and a 10 day course of CsA from BN whole blood by density gradient centrifugationusing Lympholyte-Rat (density, 1.094 g/ml; Cedarlane,(10 mg/kg/day intramuscularly). Ten days after cell infu-

sion, the animals underwent kidney transplant from the Ontario, Canada). The PMN-enriched preparation wasobtained by centrifugation of whole blood on Lympho-same BN donor strain, and CsA was discontinued. Simi-

larly, an additional group of Lewis rats (group 14, N 5 4) lyte-Mammal (density, 1.086 g/ml; Cedarlane). The pelletcontaining PMNs and red blood cells was treated astreated with the same short course of CsA but without

BN leukocytes received a BN kidney allograft 10 days described for total leukocyte preparation.Trypan blue exclusion testing always showed a moreafter starting the immunosuppressants.

We then investigated whether a particular cell popula- than 90% cell viability before intravenous injection intothe penile vein of the Lewis recipient. The cell count oftion, among peripheral blood leukocytes, may be respon-

sible for the potential tolerogenic effect of this cell infu- leukocyte suspension was determined to allow a uniformcell dose and therefore an adequate comparison of thesion. Lewis rats were intravenously given BN PBMCs

(70 3 106 per experiment, N 5 6, group 15) or an enriched results.preparation of BN polymorphonuclear cells (PMNs;

Mixed lymphocyte reaction30 3 106 per experiment, N 5 3, group 16) under thesame immunosuppression regimen with CsA, as de- Peripheral blood mononuclear cells were isolated

from Lewis rats on density gradient using the Lympho-scribed earlier in this article. This number of cells reflectsthe relative contribution of the two populations to the lyte Rat. The latter cells (6 3 105) were mixed with 6 3

105 allogeneic spleen cells from BN rats that had beentotal number of 100 3 106 peripheral blood leukocytes(70% PBMCs and 30% PMNs) infused in the first set irradiated (4000 rad) in vitro and cultured in tissue cul-

ture tubes in 1.2 ml culture medium, RPMI-1640 supple-of experiments. Forty days after cell infusion, the animalswere transplanted with a BN kidney, and graft survival mented with 2 mm l-glutamine, 100 mg/ml streptomycin,

100 U/ml penicillin, 0.05 mm 2-mercaptoethanol, 1 mmwas monitored thereafter.pyruvic acid, and 20% fetal bovine serum (FBS). After

Kidney transplantation three days of culture, 200 ml of each tube were trans-ferred in triplicate in 96-well round-bottomed plates andOrthotopic kidney transplantation was performed as

previously described [19]. BN donor animals were anes- pulsed for 18 hours with 3H-thymidine (1 mCi/well;Amersham International, Buckinghamshire, UK). Cellsthetized with leptofen (1 mg/kg i.m.). The left kidney

was prepared by freeing the ureter from the attachments. were then harvested, and their 3H-thymidine uptake wasmeasured with a b counter. In each experiment, controlThe renal artery was separated from the renal vein by

blunt dissection. The donor kidney and ureter were re- combinations of Lewis PBMCs and irradiated Lewisspleen cells were performed. The stimulation index (SI)moved en bloc and flushed with normal saline solution

containing 1000 UI/ml heparin. Then the kidney was was calculated using the following formula:

Noris et al: Donor leukocytes and kidney allograft survival1104

SI 5CPM in allogeneic MLR

CPM in control combinations

where CPM is counts per minute.

Analysis of chimerism

The chimerism was first evaluated by flow cytometryassay using a donor-specific MHC class I antibody. Pe-ripheral blood was obtained from the tail vein of Lewisrats into a heparinized syringe, and leukocytes in wholeblood labeled with the monoclonal antibody from OX27clone (Serotec, Oxford, UK), which recognizes rat MHCclass I RT1n (BN) but not RT1l (Lewis). Washed cellswere then stained with FITC-conjugated secondary anti-body to mouse IgG precleared for rat IgG (Serotec),treated with ammonium chloride to lyze red blood cells,and analyzed by flow cytometry using a FACSort appara-tus (Becton Dickinson, San Jose, CA, USA). Examiningperipheral leukocytes of Lewis rats given in addition toBN leukocytes a short course of CsA, positive cells werefound up to 20 days after cell infusion. However, wealso observed that in Lewis rats given the same CsAtreatment but not BN donor cells, the peripheral leuko-cytes were also stained for BN MHC class I antigen,indicating an artifact induced by the drug that probablyenhances the unspecific binding of both primary andsecondary antibodies (Fig. 1). In contrast, peripheral leu-kocytes from untreated Lewis rats were never stainedby OX27 antibody.

The peripheral blood chimerism was then assessedwith polymerase chain reaction (PCR) by using specificprimers for the donor BN MHC class II RT1.Bbn. Be-cause only few short fragments of MHC class II of BNrats were reported in gene banks, we began by cloningthe highly polymorphic RT1.Bbn exon 2. A 336 bp frag-ment containing exon 2 was amplified from genomic BNDNA by designing PCR primers homologous to evo-lutionary-conserved mouse intron sequences flankingthe exon [20]. The amplified product was ligated intopCR2.1 vector (TA cloning; Invitrogen, Leek, The Neth-erlands) and sequenced by dideoxynucleotide sequenc-ing (Sequenase Version 2.0; United States Biochemical,Cleveland, Ohio, USA). To eliminate potential errors Fig. 1. FACS profile of Lewis peripheral blood lymphocytes stained

for major histocompatibility complex (MHC) class I as described in theintroduced by PCR, several independent clones wereMethods section. Lymphocyte population was recognized on the basis ofanalyzed. The obtained sequence overlapped with a pre-its light scattering parameters. Symbols are: (open histogram) secondary

viously reported RT1.Bbn fragment (GenBank M76781; antibody alone; (gray histogram) MHC class I staining. (A) ControlLewis rat. (B) Lewis rat 10 days after donor Brown-Norway (BN)Fig. 2) from base 98 to 226, with two nucleotide differ-leukocyte infusion and CsA treatment. (C) Lewis rat after 10 days CsAences at bases 140 and 195. Alignment of BN sequencestreatment.

with the correspondent Lewis sequence RT1.Bbl isshown in Figure 2.

For microchimerism analysis, blood was collectedfrom the rat tail vein on ethylenediaminetetraacetic acid trophotometer, aliquoted, and stored at 48C until an-(EDTA)-containing tubes. Genomic DNA was isolated alysis. Specific primers (59-CGCAGGGGATTTCGTfrom buffy coats with a DNA extraction kit (Nucleon ATT-39, p1; 59-GGTGGGGACCTCCGTCT-39, p2) were

designed on the basis of the BN RT1.Bbn sequence ob-BACC 2; Amersham), quantitated on an ultraviolet spec-

Noris et al: Donor leukocytes and kidney allograft survival 1105

Fig. 2. Nucleotide sequence of exon 2 (capi-tal letters) and partial intron sequences (smallletters) of MHC class II RT1.Bbn obtainedby cloning. This sequence was aligned withRT1.Bbl sequence (X56596) and with a pre-viously described fragment of RT1.Bbn exon2 (M76781). Underlined are the sequence ofprimers 1, 2, and 3 (p1, p2, p3) used for mi-crochimerism evaluation. Only different nu-cleotides are shown in the figure, and identicalnucleotides are indicated by dashes. This se-quence is available from GenBank under ac-cession number AF 113922.

tained by cloning. The PCR reactions were done with identical except for the initial hot start (10 min) and forthe number of cycles that was reduced to 32. In these100 ng of the DNA template, 18 pmol of forward and

reverse primers, 200 mm dNTPs, 1.5 mm MgCl2, 1 3 experimental conditions, Lewis DNA did not amplify.PCR products were analyzed by electrophoresis in 1%GeneAmp II PCR buffer (10 mm Tris HCl, pH 8.3, 50

mm KCl), and 1 U AmpliTaqGold (Perkin-Elmer, NJ, agarose gel with 1 mg/ml ethidium bromide added forultraviolet visualization.USA) in a final volume of 20 ml. The PCR reaction was

performed in a Perkin-Elmer 9700 thermal cycler byStatistical analysis60 cycles of denaturation (948C, 45 seconds), annealing

(588C, 30 seconds), and extension (728C, 30 seconds) Statistical analysis of kidney graft survival was per-with a pre-PCR heat step of five minutes. This method formed using PROC LIFETEST of SAS 6.12, and sig-specifically detected BN DNA (PCR product 256 bp) at nificance of difference between groups was determineda 1:1000 dilution in Lewis DNA. No signal was found by log-rank test. Analysis of MLR results was performed

by one-way analysis of variance (ANOVA). Statisticalwith DNA from naive Lewis rats.significance was defined as P , 0.05.By a second PCR round, BN DNA could be specifi-

cally detected until the dilution was larger than 1:2500in the presence of Lewis DNA. One microliter of the

RESULTSfirst PCR product, previously diluted 1:200 in water, wasDonor peripheral blood leukocyte infusion and kidneyused as a template, and a more internal forward primergraft survivalwas used in combination with the same reverse primer

of the first amplification (59-ACGCAGCGCATACG Table 1 shows kidney allograft survival in Lewis ratsreceiving BN donor peripheral blood leukocyte infusionGCTC-39, p3; 59-GGTGGGGACCTCCGTCT-39, p2;

PCR product 204 bp). The PCR reaction conditions were at different time points before transplantation. One out

Noris et al: Donor leukocytes and kidney allograft survival1106

Table 1. Effect of pretransplant donor peripheral blood leukocyte (group 7) rejected their graft within nine days after trans-infusion at different times before transplantation on Brown-

plantation. Seventy-five percent of animals pretreatedNorway (BN) kidney allograft survival in Lewis ratswith the third-party WF peripheral blood leukocyte infu-

Time of BN leukocyte infusion Kidney graft survival sion (group 8) had only a minor prolongation of BNGroups days pre-transplant dayskidney graft survival, up to 25 days postsurgery (P 5

Group 1 Day 60 7, 7, 8, 670.014 vs. group 9, P 5 0.0084 vs. group 7, P 5 0.0034Group 2 Day 40 7, 7, 7, 7, 24

Group 3 Day 30 6, 6, 6, 7, 12, 56, .70 vs. group 6). Untreated control animals rejected theGroup 4 Control (no cell infusion) 7, 8, 13 allograft within nine days of their kidney transplant

(group 9).The long-term surviving animals pretreated with BN

donor leukocyte preparation (group 5) had a normal orof four rats (25%) given the donor leukocyte preparation moderately impaired renal function 60 days after trans-60 days before transplantation had a graft survival that plantation (serum creatinine, range 0.89 to 2.35 mg/dl).was prolonged up to the 67th postoperative day. The In the same group, those rats that developed graft rejec-remaining transplanted animals rejected their grafts tion had elevated serum creatinine values at the timewithin eight days postsurgery. Similarly, infusion of do- of sacrifice (range 1.83 to 5.55 mg/dl). In all controlnor leukocytes 40 days before kidney transplant did not rats given CsA alone, graft failure was associated withprovide any substantial benefit, because the graft sur- elevated serum creatinine values (range 6.23 to 7.60vived up to 24 days postsurgery in only one rat (20%). mg/dl). The same occurred in transplanted animals re-Further reduction in the interval between leukocyte infu- ceiving the third-party WF kidney graft (group 7), andsion and kidney transplant to 30 days prolonged graft their serum creatinine concentration at the time of graftsurvival in one rat (day 56) and induced tolerance in failure ranged from 3.61 to 6.06 mg/dl. In group 8 trans-another one (more than 70 days; 28.5%), but all of the planted animals given the third-party WF leukocyteremaining animals rejected their graft within 12 days preparation, serum creatinine ranged from 1.97 to 2.50postsurgery. In control rats given a kidney transplant mg/dl.alone, the graft failed within the 13th postoperative day. In long-term surviving rats of group 5, systemic unre-

sponsiveness was established by evaluating the func-Effect of combined immunosuppression and donor tional status of T cells on the MLR in vitro. Figure 4peripheral blood leukocyte infusion on shows the stimulation index (SI) of untreated Lewis ratsallograft survival as compared with that of tolerant and third-party kidney

Lewis rats, which also received short-term CsA at the graft-rejecting hosts. The SI of the untreated controltime of donor cell infusion, were healthy during the 40 responder cells to BN stimulator cells was, on average,days after cell infusion and before kidney transplant. In 16.7. By contrast, the MLR response of the tolerant hostthe same period, body weight increased from baseline cells was markedly and significantly reduced with a meanvalues of 175 to 322 g (group 5) and from 178 to 314 g SI of 2 (P , 0.05 vs. untreated controls). On the other(group 7). Similar body weight gain was observed in hand, the proliferative response of cells from Lewis ani-group 8 rats that received WF leukocytes 40 days before mals pretreated with WF leukocyte infusion and re-BN kidney allograft (175 to 320 g). Rats given CsA alone jecting a BN kidney allograft was blunted, but still fourwere similarly healthy, and their body weights (176 to times higher than that of cells from tolerant animals328 g) increased to a comparable extent as those animals (P 5 NS vs. untreated controls).given the leukocyte suspension.

Analysis of microchimerism in the peripheral blood ofFigure 3 shows kidney graft survival in these fiveLewis recipients of donor leukocytes with orgroups of animals. All rats given the infusion of periph-without cyclosporineeral blood leukocytes under the short course of CsA

(group 5) had prolonged survival of their kidney allo- To detect donor microchimerism in Lewis peripheralgrafts (median survival, 53 days; P 5 0.0001 vs. group blood, we designed a set of primers capable of detecting6, 7, 9 and P 5 0.0012 vs. group 8). Sixty-eight percent nucleotide differences for RT1.Bb genes of the donorof these animals rejected their graft within 20 to 69 days and recipient. The specificity and sensitivity of thesepost-transplant, whereas in the remaining rats a state of primers was first tested using genomic BN DNA seriallytolerance developed, as documented by a graft survival diluted with a constant amount of Lewis DNA. As shownof more than 70 days. At variance, rats treated with CsA in Figure 5, this method specifically detected BN DNAalone in the pretransplant period (group 6) invariably at more than a 1:2500 dilution in the presence of Lewisrejected kidney graft within eight days of postsurgery. DNA.Similarly, Lewis rats given BN peripheral blood leuko- Then, the blood samples from Lewis rats infused with

BN peripheral blood leukocytes were examined. Ascytes and receiving the third-party WF kidney transplant

Noris et al: Donor leukocytes and kidney allograft survival 1107

Fig. 3. Graft survival curves in Lewis ratspretreated with Brown-Norway (BN) orWistar Furth (WF) peripheral blood leuko-cytes under a short-course of cyclosporine(CsA) and receiving a BN or WF kidneytransplant. Symbols are: (s) Csa 1 BN leu 1BN kidney (group 5); (j) CsA 1 BN kid-ney (group 6); (h) CsA 1 BN leu 1 WFkidney (group 7); (d) CsA 1 WF leu 1 BNkidney (group 8); (m) untreated control(group 9); *P 5 0.0001 vs. groups 6, 7, 9; 8P 50.0012 vs. group 8; #P 5 0.014 vs. group 9;§P 5 0.0084 vs. group 7; ¶P 5 0.0034 vs.group 6.

Effect of repeated donor leukocyte infusion on kidneygraft survival

Fifty percent of the animals that received the twoinfusions of 100 3 106 BN peripheral blood leukocytesa week apart (group 12) had a prolonged kidney graftsurvival (days 27, 28, 32, and 33 post-transplant). Theremaining transplanted rats rejected their graft within 7to 14 days postsurgery. In this group of animals, graftsurvival was lower than that of transplanted Lewis ratspreviously receiving a single infusion of the leukocytepreparation, suggesting that a possible sensitization ofthe recipient may have occurred.

Effect of shortening the time interval between donorcell infusion and kidney transplant

Prolongation of kidney allograft survival was achievedafter reducing the time interval between BN peripheralFig. 4. Stimulation index of mixed lymphocyte reaction (MLR) of re-blood leukocyte infusion and the transplant from thesponder cells from untreated control Lewis rats (N 5 5), tolerant Lewis

rats (N 5 4), or Lewis rats pretreated with WF leukocytes and rejecting same donor rat strain to 10 days. Rats received a 10 daya BN graft (N 5 3), to BN stimulator cells. Tolerant responder cells course of CsA starting on the day of cell infusion (groupwere harvested from Lewis hosts with BN kidney allograft surviving

13). In three out of five animals, graft survival was pro-for more than 70 days. Values are mean 6 sd. *P , 0.05 vs. untreatedcontrols. longed more than 50 days post-transplant, whereas in

the remaining two rats, the kidney allograft failed at 17and 30 days postsurgery (Table 2). In all Lewis rats givenCsA but not BN leukocyte infusion and transplanted 10shown in Figure 5, in all animals not receiving the con-days after starting the short-course of the immunosup-comitant immunosuppression with CsA (group 11), BNpressive agent (group 14), the BN kidney graft failedDNA products in the peripheral blood were demon-within 10 days postsurgery (P 5 0.004 vs. group 13).strated for up to two days after donor cell infusion. After

the second PCR round, a band was also observed at 10Effect of donor peripheral blood mononuclear cell orand 20 days, whereas no signal was ever found at 40polymorphonuclear cell infusion on kidneydays. On the other hand, Lewis rats given BN leukocytesgraft survivaland the short course of CsA (group 10) demonstrated

Table 3 reports BN allograft survival in Lewis ratsthe continuous presence of donor BN DNA until 40 daysrecipients of BN PBMC (group 15) or BN PMN (groupafter cell infusion (Fig. 5). Microchimerism persisted in16) cell infusion under concomitant short-term immuno-four out of six long-term graft recipients 60 to 70 days

post-transplantation. suppression with CsA. Sixty-seven percent of the animals

Noris et al: Donor leukocytes and kidney allograft survival1108

Fig. 5. Microchimerism in Lewis rats receiving peripheral leukocytes from BN rats as evaluated by PCR using BN-specific primers. (A) Oneround of PCR with primers p1 and p2. (B) Two rounds of PCR, primers p1 and p2, followed by p3 and p2. (Standard) Titration of the specificityand sensitivity of the PCR for BN rat DNA. DNAs from BN and Lewis blood were mixed at ratios ranging from 1:10 to 1:2500. Lewis DNA wasused as negative control. LW, Lewis. (CsA) Sequential study (1 to 40 days postinfusion) of microchimerism in peripheral blood of Lewis ratsreceiving 100 3 106 BN leukocytes intravenously and a concomitant short course CsA treatment (discussed in Experimental design section).Peripheral blood from BN and Lewis rats was used as positive or negative control, respectively. (No CsA) Sequential study (1 to 40 days postinfusion)of microchimerism in peripheral blood of Lewis rats receiving 100 3 106 BN leukocytes intravenously but no CsA. Peripheral blood obtainedfrom BN and Lewis rats was used as positive or negative control, respectively. (Long-term) Detection of microchimerism in peripheral blood fromlong-term Lewis recipients of a BN renal graft. Forty days before transplantation, recipient rats received BN leukocytes and a short course ofCsA as described in the Experimental design section. Peripheral blood was obtained 60 to 70 days after transplantation. A specific positive signalwas found in four out of six animals after a two-round PCR. Animals are identified by a number. Results are representative of four independentexperiments.

Table 3. Effect of pretransplant infusion of donor peripheral bloodTable 2. Effect of pretransplant infusion of donor peripheral bloodleukocytes 10 days before transplantation on Brown-Norway mononuclear cells or polymorphonuclear cells on Brown-Norway

kidney allograft survival in Lewis ratskidney allograft survival in Lewis rats

Time of BN leukocyte Time of BN leukocyteinfusion Kidney graft survivalinfusion Kidney graft survival

Groups days pre-transplant days Groups days pre-transplant days

BN-PBMC Day 40 19, 55, .70, .70,CsA alone — 8, 9, 10, 10BN-Leu 1 CsA Day 10 17, 30, 58, .70, .70a .70, .70a

BN-PMN Day 40 13, 13, 32aAbbreviations are: BN, Brown-Norway rats; Leu, peripheral blood leukocytes;

CsA, cyclosporine (10 mg/kg/day i. m., for 10 days). Abbreviations are: BN, Brown-Norway rats; PBMC, peripheral blood mono-a P 5 0.004 vs. CsA alone nuclear cells; PMN, polymorphonuclear cells.

a P 5 0.012 vs. BN-PMN group

that received donor PBMC infusion had an indefinite to 1.99 mg/dl). By contrast, Lewis rats given donor PMNsurvival of their kidney graft from the same BN donor cell infusion invariably rejected a subsequent BN kid-strain and had a quite normal renal function after 70 ney allograft, with mild prolongation of graft survival

achieved in only one recipient (P 5 0.012 vs. group 15).days post-transplantation (serum creatinine range, 1.56

Noris et al: Donor leukocytes and kidney allograft survival 1109

DISCUSSION All together, the above data extend previous observa-tions that a Lewis donor bone marrow cell suspensionThis study shows that a single intravenous infusion ofunder a short course of tacrolimus induced an indefinitedonor peripheral blood leukocytes from BN rats intosurvival of Brown Norway liver or heart transplantedMHC-incompatible Lewis recipients largely failed to100 days after donor cell infusion [18]. Our findings alsoprolong a subsequent kidney allograft from the sameexplain why in previous rat studies donor leukocytesdonor transplanted 60, 40, or 30 days after cell infusion.injected at the time of cardiac transplant with no concom-The failure to observe consistent prolongation of kidneyitant immunosuppression failed to prolong heart survivalgraft survival with the pretransplant donor leukocyte[17]. Indeed, a short course of immunosuppression, asinfusion alone can be due to an insufficient antigen loadused in this study, seems necessary to allow enough do-provided with 100 3 106 BN cells. However, previousnor antigen to be presented to host cells to activate thestudies have found that incompatible rats given donortolerogenic signals that ultimately lead to the deletionliver or spleen leukocytes at the time of grafting rejectedor functional inactivation of responder T cells [9, 22].the transplanted heart when even greater doses of cells

On this line of investigation, we also found that thewere infused [17]. Alternatively, the lack of donor leuko-short course of CsA treatment is mandatory to allowcyte engraftment may explain the failure to prolong kid-engraftment of donor BN leukocytes, as documented byney allograft survival. Starzl et al have indicated thatthe persistent microchimerism by PCR up to 40 daysengraftment of donor bone marrow cells depends on theafter cell infusion in animals given a concomitant immu-adequacy of concomitant immunosuppression given tonosuppression with CsA, but not in those that did notprotect donor leukocytes from host-reactive T cells [21].receive the drug.According to Starzl et al, an ideal interaction of donor

It has been suggested that actively acquired toleranceand recipient leukocytes is required to establish a veryis concordant with donor lymphomyeloid cell chimerism,low-grade stimulatory state, which is necessary for recip-since the early discovery of a classic neonatal tolerancerocal nonreactivity among the two cell populations [6,[9, 23]. In most long-term graft recipients, we demon-

7, 9, 21]. Taking advantage of this experience, we studiedstrated peripheral blood microchimerism up to 60 to 70

the effect of donor leukocyte infusion on subsequent days post-transplantation. This suggests that the persis-kidney allograft survival in rats given some immunosup- tence of donor cells in the host may be important topressants at the time of cell infusion. Thus, in these drive the sequence of events that ultimately maintainsexperiments, a short course of CsA was started on the donor-specific graft unresponsiveness [24, 25]. However,same day of infusion of the donor leukocyte preparation. controversies remain as to whether the cellular microchi-Under this treatment condition, prolongation of graft merism is the cause or rather the result of graft accep-survival was achieved when the kidney transplant was tance in long-term recipients [26–28].performed 40 days later, with some animals even devel- Failure to induce permanent tolerance in 100% ofoping tolerance to the allograft. By contrast, CsA pre- animals could possibly depend on the initial donor leuko-treatment alone did not affect graft survival. Our findings cyte load, which might not have been enough to reachsuggest that the short course of CsA treatment allowed the threshold to fully drive a tolerogenic signal. Recentan engraftment of donor leukocytes. They also indicate findings in rodents and humans found that the successthat these cells possess more potential for enhancing the of donor bone marrow infusion could only be achievedinherent, but low tolerogenic potential of donor kidney when a high enough dose of donor marrow-derived stemrelated to its smaller “passenger leukocyte” components cells was used [17, 29].than other whole organs such as liver [21]. This effect We also found that repeated-donor leukocyte infu-was donor specific because a third-party kidney allograft sions under concomitant immunosuppression with CsAwas invariably rejected. However, when third-party pe- not only failed to achieve tolerance, but even sensitizedripheral blood leukocytes were infused, some prolonga- the graft recipients to the extent that kidney graft sur-tion of BN kidney allograft survival was also observed. vival was actually shorter than what we obtained with theThis unexpected finding is difficult to interpret. One pos- pretransplant infusion of a single leukocyte preparation.sibility is that WF leukocytes may share minor histo- This is in line with an earlier clinical observation that acompatibility antigens with BN kidneys or leukocytes, protocol of repeated pretransplant donor-specific bloodwhereas this is not the case for BN leukocytes and WF transfusion is associated with sensitization of the recipi-kidney. This is in line with the in vitro observation that ent and may preclude transplantation in near 14 to 30%the MLR response of Lewis cells from animals pre- of the donor-recipient pairs [30–32].treated with WF leukocytes and receiving BN kidney In this study, kidney grafts were first performed onwas not as high as that of lymphocytes from untreated day 40 after the infusion of donor leukocytes in the LewisLewis rats, but was still higher than that of responder recipient. However, similar results have been achieved

by transplanting kidneys 10 days after donor leukocytecells from tolerant rats.

Noris et al: Donor leukocytes and kidney allograft survival1110

infusion, always depending on short-term CsA. This mocytes bearing the specific donor T-cell receptor. Theseresults are consistent with the interpretation that intra-would open interesting perspectives of shortening the

interval between donor cell infusion and the subsequent thymic clonal deletion is the only significant mechanismby which tolerance is maintained in these mixed chimerastransplant, keeping in mind future clinical developments.

The heterogeneity of the leukocyte population infused [42]. In line with these findings are also previous observa-tions by our group and other investigators that directin the host before kidney transplantation precludes

definite conclusions on the properties of prolonging graft injection of donor cells/antigens into the thymus is capa-ble of inducing donor-specific tolerance to a subsequentsurvival afforded per se by any specific cell lineages in

the preparation. In additional experiments, however, we tissue or vascularized organ allograft through activationof a process of programmed cell death/apoptosis of ma-documented that donor PBMCs but not PMNs in the

preparation contributed to prolong kidney allograft sur- turing host thymocytes [44–46].All together, these results are consistent with the possi-vival, with tolerance achieved in most animals. This is

in line with previous findings in a model of rat small bility that after systemic infusion, donor mononuclearcells and few hematopoietic precursor cells migrate tobowel transplantation in which prolongation of graft sur-

vival was achieved by donor splenocyte infusion via the the thymus shortly thereafter and become capable ofnegatively selecting newly developing T cells. Eventu-portal vein under a short course of the immunosuppres-

sive agent FK506 [33]. Which one is the specific donor ally, the peripheral T-cell components would be devoidof alloreactive cell populations. This mechanism maycell lineage that is actively involved remains ill defined.

Very recently, it has been shown in mice that T cells operate in concert with veto cells present in the donorperipheral blood leukocytes providing a reliable meanfrom the spleen of the donor facilitate the acceptance

of the allogeneic skin engraftment when injected into of ensuing donor- and host-specific unresponsiveness.Thus, pretransplant donor leukocyte infusion underthe portal vein of the recipients, an effect that was po-

tentiated when donor bone marrow cells were also given the appropriate conditions tips the immune balance to-ward improved graft acceptance. The tolerance potentialintravenously [34]. Thus, the possibility exists that in our

experimental conditions, the donor T-cell component of of such treatment is a further step toward the ultimategoal of inducing donor-specific tolerance to allografts,the mononuclear cell population we infused into the

recipient may have played an important role in pro- leaving intact the response to third-party antigens.This approach has obvious implications for clinicallonging the survival of the subsequent kidney allograft.

Calne and Davies have suggested a critical role of class transplantation.I antigens released by donor cells in the circulation forfacilitating the achievement of the tolerant state [35, 36]. ACKNOWLEDGMENTSCirculating MHC class I molecules may neutralize donor This work has been partially supported by a European Community

Grant (PL 970525). This work was submitted as an abstract to theclass I-specific, primed cytotoxic T cells in the recipient’s18th Annual Scientific Meeting of the American Society of Transplantperiphery. However, purified soluble class I MHC mole-Physicians (May 15–19, 1999). We thank Dr. Susana C. Amuchastegui

cules had minimal [37] or no immunosuppressive or tol- for her contribution in performing some preliminary experiments. Weare also grateful to Dr. Roberto Benini for his help in performingerogenic [38, 39] effect in rat models of liver or kidneystatistical analysis.allotransplantation.

The presence of cells with a “veto” function has been Reprint requests to Dr. Marina Noris, Mario Negri Institute for Phar-macological Research, Via Gavazzeni 11, 24125 Bergamo, Italy.proposed as an alternative mechanism for the inductionE-mail: noris@irfmn.mnegri.itof donor-specific tolerance [40]. Donor bone marrow-

derived cells with “veto” function have been identifiedin the PBMCs of long-term kidney and bone marrow- APPENDIXtransplanted nonhuman primates [41]. Thus, “veto” cells

Abbreviations used in this article are: BN, Brown Norway rats;could be part of the leukocyte preparation and could CsA, cyclosporine A; CPM, counts per minute; GVHD, graft versus

host disease; LW, Lewis rats; MHC, major histocompatibility complex;contribute to a down-regulation of the host immune re-MLR, mixed lymphocyte reaction; PBMC, peripheral blood mononu-sponse in a donor-specific manner.clear cells; PCR, polymerase chain reaction; PMN, polymorphonuclear

Another intriguing possibility to explain the tolero- cells; WF, Wistar Furth rats.genic effect of donor leukocytes derives, however, fromstudies in mixed allogeneic bone marrow chimeras in REFERENCESmice prepared with a nonmyeloablative-conditioning

1. Charlton B, Auchincloss H, Wood M, Sahyoun A, Codish S,regimen [15, 42, 43]. In these mice, long-term donor- Brown R: Mechanisms of transplantation tolerance. Annu Rev

Immunol 12:707–734, 1994specific skin graft tolerance [15] was associated with the2. Brennan D, Mohanakumar T, Flye W: Donor-specific transfusionearly presence of donor class II-positive cells with den-

and donor bone marrow infusion in renal transplantation tolerance:dritic morphology in the thymus, and the presence of A review of efficacy and mechanisms. Am J Kidney Dis 26:701–715,

1995these cells correlated with deletion of mature host thy-

Noris et al: Donor leukocytes and kidney allograft survival 1111

3. Barber H, Mankin J, Laskow D, Deierhoi M, Julian B, Curtis 23. Billingham RE, Brent L, Medawar PB: Actively acquired toler-ance of foreign cells. Nature 172:603–606, 1953J, Diethelm A: Long-term results of a controlled prospective study

with transfusion of donor-specific bone marrow in 57 cadaveric 24. Ehl S, Aichele P, Ramseier H, Barchet W, Hombach J, PircherH, Hengartner H, Zinkernagel R: Antigen persistence and timerenal allograft recipients. Transplantation 51:70–75, 1991

4. Fontes P, Rao A, Demetris A, Zeevi A, Trucco M, Carroll P, of T-cell tolerization determine the efficacy of tolerization proto-cols for prevention of skin graft rejection. Nature Med 4:1015–1019,Rybka W, Rudert W, Ricordi C, Dodson F, Shapiro R, Tzakis

A, Todo S, Abu-Elmagd K, Jordan M, Fung J, Starzl T: Bone 199825. Starzl T, Demetris A, Murase N, Trucco M, Thomson A, Raomarrow augmentation of donor-cell chimerism in kidney, liver,

heart, and pancreas islet transplantation. Lancet 344:151–155, 1994 A: The lost chord: Microchimerism and allograft survival. ImmunolToday 17:577–584, 19965. Ricordi C, Karatzas T, Nery J, Webb M, Selvaggi G, Fernandez

H, Bean J, Esquenazi V, Joshua M, Tzakis A: High-dose donor 26. Yokoi Y, Hirasawa S, Iwaya M, Okuyama S, Nakamura S, BabaS, Miyamoto M, Tsujimoto G, Amemiya H, Kimura H: Microchi-bone marrow infusions to enhance allograft survival: The effect

of timing. Transplantation 63:7–11, 1997 merism and liver graft acceptance. Transplant Proc 27:1555–1557,19956. Starzl T, Demetris A, Trucco M: Cell migration and chimerism

after whole-organ transplantation: The basis of graft acceptance. 27. Suberbielle C, Caillat-Zucman S, Legendre C, Bodemer C,Noel LH, Kreis H, Bach JF: Peripheral microchimerism in long-Hepatology 17:1127–1152, 1993

7. Starzl T, Demetris A, Murase N, Ildstad S, Ricordi C, Trucco term cadaveric-kidney allograft recipients. Lancet 343:1468–1469,1994M: Cell migration, chimerism, and graft acceptance. Lancet

339:1579–1582, 1992 28. Schlitt HJ, Hundrieser J, Hisanaga M, Uthoff K, Karck M,Wahlers T, Wonigeit K, Pichlmayr R: Patterns of donor-type8. Rao A, Thomson A, Shapiro R, Starzl T: Chimerism after whole

organ transplantation: Its relationship to graft rejection and toler- microchimerism after heart transplantation. Lancet 343:1469–1471,1994ance induction. Curr Opin Nephrol Hypertens 3:589–595, 1994

9. Starzl T, Zinkernagel R: Antigen localization and migration in 29. Reisner Y, Martelli M: Bone marrow transplantation acrossHLA barriers by increasing the number of transplanted cells. Im-immunity and tolerance. N Engl J Med 339:1905–1913, 1998

10. Thomson AW, Lu L, Subbotin V, Starzl T: Propagation of den- munol Today 16:437–440, 199530. Salvatierra O Jr, Vincenti F, Amend W, Potter D, Iwaki Y,dritic cell progenitors from mouse liver and their in vivo migration

to T-dependent areas of allogeneic lymphoid tissue. Transplant Opelz G, Terasaki P, Duca R, Cochrum K, Hanes D, Stoney RJ,Feduska NJ: Deliberate donor-specific blood transfusions prior toProc 26:3484–3486, 1994

11. Thomson AW, Lu L, Subbotin VM, Li Y, Qian S, Rao AS, Fung living related renal transplantation: A new approach. Ann Surg192:543–552, 1980JJ, Starzl TE: In vitro propagation and homing of liver-derived

dendritic cell progenitors to lymphoid tissues of allogeneic recipi- 31. Salvatierra OJ, Melter J, Potter D, Garavoy M, Vincenti F,Amend W, Husing R, Hopper S, Feduska N: A 7-year experienceents: Implications for the establishment and maintenance of donor

cell chimerism following liver transplantation. Transplantation with donor specific blood transfusions. Transplantation 40:654–659,198559:544–551, 1995

12. Steinman RM, Inaba K, Austyn JM: Donor-derived chimerism 32. Colombe B, Lou C, Salvatierra OJ, Garovoy M: Two patternsof sensitization demonstrated by recipients of donor-specific trans-in recipients of organ transplants. Hepatology 17:1153–1156, 1993

13. Terakura M, Murase N, Demetris A, Ye Q, Thomson A, Starzl fusion. Limitations to control by Imuran. Transplantation 44:509–515, 1987T: Lymphoid/nonlymphoid compartmentalization of donor leuko-

cyte chimerism in rat recipients of heart allografts, with or without 33. Miyauchi T, Ishikawa M, Tashiro S, Hisaeda H, Nagasawa H,Himeno K: Effect of donor-specific splenocytes via portal veinadjunct bone marrow. Transplantation 66:350–357, 1998

14. Ildstad ST, Sachs DH: Reconstitution with syngeneic plus alloge- and FK506 in rat small bowel transplantation. Transplantation65:27–32, 1998neic or xenogeneic bone marrow leads to specific acceptance of

allografts or xenografts. Nature 307:168–170, 1984 34. Morita H, Sugiura K, Inaba M, Jin T, Ishikawa J, Lian Z, AdachiY, Sogo S, Yamanishi K, Taki H, Adachi M, Noumi T, Kamiyama15. Tomita Y, Khan A, Sykes M: Role of intrathymic clonal deletion

and peripheral anergy in transplantation tolerance induced by bone Y, Good R, Ikehara S: A strategy for organ allografts withoutusing immunosuppressants or irradiation. Proc Natl Acad Sci USAmarrow transplantation in mice conditioned with a nonmyeloabla-

tive regimen. J Immunol 153:1087–1098, 1994 95:6947–6952, 199835. Calne R: New strategies in tolerance. Med Sci Res 19:230–231,16. De Vries-van der Zwan A, Besseling A, de Waal L, Boog

C: Specific tolerance induction and transplantation: A single-day 199136. Calne R, Davies H: Organ graft tolerance: The liver effect. Lancetprotocol. Blood 89:2596–2601, 1997

17. Sun J, Sheil A, Wang C, Wang L, Rokahr K, Sharland A, 343:67–68, 199437. Sumimoto R, Kamada N: Specific suppression of allograft rejectionJung S-E, Li L, McCaughan G, Bishop G: Tolerance to rat liver

allografts. IV. Acceptance depends on the quantity of donor tissue by soluble class I antigen and complexes with monoclonal antibody.Transplantation 50:678–682, 1990and on donor leukocytes. Transplantation 62:1725–1730, 1996

18. Murase N, Starl T, Tanabe M, Fujisaki S, Miyazawa H, Ye Q, 38. Spencer SC, Fabre JW: Bulk purification of a naturally occurringsoluble form of RT1-A class I major histocompatibility complexDelaney C, Fung J, Demetris A: Variable chimerism, graft-versus-

host disease, and tolerance after different kinds of cells and whole antigens from DA rat liver, and studies of specific immunosuppres-sion. Transplantation 44:141–148, 1987organ transplantation from Lewis to Brown Norway rats. Trans-

plantation 60:158–171, 1995 39. Priestley CA, Dalchau R, Sawyer GJ, Fabre JW: A detailedanalysis of the potential of water-soluble classical class I MHC19. Perico N, Imberti O, Bontempelli M, Remuzzi G: Toward novel

antirejection strategies: In vivo immunosuppressive properties of molecules for the suppression of kidney allograft rejection and invitro cytotoxic T cell responses. Transplantation 48:1031–1038, 1989CTLA4Ig. Kidney Int 47:241–246, 1995

20. She J, Boehme S, Wang T, Bonhomme F, Wakeland E: Amplifica- 40. Thomas J, Verbanac K, Thomas F: The veto mechanism in trans-plant tolerance. Transplant Rev 5:209–229, 1991tion of major histocompatibility complex class II gene diversity by

intraexonic recombination. Proc Natl Acad Sci USA 88:453–457, 41. Thomas J, Carver F, Kasten-Jolly J, Haisch C, Rebellato L,Gross U, Vore S, Thomas F: Further studies of veto activity in1991

21. Starzl T, Demetris A, Murase N, Thomson A, Trucco M, Ri- Rhesus monkey bone marrow in relation to allograft toleranceand chimerism. Transplantation 57:101–115, 1994cordi C: Donor cell chimerism permitted by immunosuppressive

drugs: A new view of organ transplantation. Immunol Today 42. Khan A, Tomita Y, Sykes M: Thymic dependence of loss of toler-ance in mixed allogeneic bone marrow chimeras after depletion14:326–332, 1993

22. Lasalle J, Hafler M: T cell anergy. FASEB J 8:601–608, 1994 of donor antigen. Transplantation 62:380–387, 1996

Noris et al: Donor leukocytes and kidney allograft survival1112

43. Nikolic B, Lei H, Pearson D, Sergio J, Swenson K, Sykes M: Deletion of donor reactive T lymphocytes in adult mice after intra-thymic inoculation with lymphoid cells. Transplantation 55:871–Role of intrathymic rat class II cells in maintaining deletional

tolerance in xenogeneic rat mouse bone marrow chimeras. Trans- 877, 199346. Jones N, Fluck N, Roelen D, Mellor A, Morris P, Wood K:plantation 65:1216–1224, 1998

44. Remuzzi G, Perico N, Carpenter C, Sayegh M: The thymic way Deletion of alloantigen-reactive thymocytes as a mechanism ofadult tolerance induction following intrathymic antigen administra-to transplantation tolerance. J Am Soc Nephrol 5:1639–1646, 1995

45. Markmann J, Odorico J, Bassiri I, Desai N, Kim J, Barker C: tion. Eur J Immunol 27:1591–1600, 1997