Proc. Nati. Acad. Sci. USAVol. 88, pp. 3218-3222, April 1991Immunology

Transplantation of wheat germ agglutinin-positive hematopoieticcells to prevent or induce systemic autoimmune disease

(stem cells/bone marrow transplantatdon/autoimmunlty)

EVELIO E. SARDIRA*, KIKUYA SUGIURA*t1, SUSUMU IKEHARAtt, AND ROBERT A. GOOD*§*Department of Pediatrics, All Children's Hospital, 801 Sixth Street South, St. Petersburg, FL 33701; and tFirst Department of Pathology and *Department ofImmunology, Kansai Medical University, Osaka, Japan 570

Contributed by Robert A. Good, January 4, 1991

ABSTRACT Hematopoietic stem cell defects are thoughtto be involved in the etiopathogenesis of systemic autoimmunedisease. Positively selected, stem cell-enriched populations ofwheat germ agglutinin-positive (WGA+) low-density bone mar-row and fetal liver cells from normal and autoimmune-pronemice were used to determine whether reciprocal transplanta-tion of stem cells between normal and autoimmune-prone miceinhibits or causes development of autoimmune disease. NZBrecipients ofDBA/2 stem cell populations analyzed >100 daysafter bone marrow or fetal liver cell transplantation showeddecreased levels of anti-DNA antibodies and decreased glomer-ular lesions when compared with nontreated NZB mice or NZBrecipients of NZB stem cell preparations. Female DBA/2recipients of WGA+ NZB bone marrow cell or fetal liver celltransplants exhibited elevated serum autoantibody levels anddeveloped glomerular lesions characteristic ofNZB mice whenanalyzed >100 days after transplantation. These pathologicaldisturbances were not observed in DBA/2 recipients ofDBA/2stem cell preparations. The data indicate thatWGA+ stem cellsfrom autoimmune-prone NZB mice contain the genetic defectsresponsible for the development of systemic autoimmune dis-ease.

Several investigators recently have described the separationof murine hematopoietic stem cells from among the cellspresent in hematopoietic tissues such as bone marrow andfetal liver (1-5). The availability of cell preparations greatlyenriched with hematopoietic stem cells has facilitated studyof the role of stem cells in immunological and hematologicaldisease. Using established cell-transfer technique, we exam-ined whether the development of systemic autoimmune dis-ease in NZB mice could be prevented by using a combinationof lethal total-body irradiation (TBI) and i.v. transplantationof positively selected stem cell-enriched preparations ofbonemarrow cells (BMC) or fetal liver cells (FLC) from normalDBA/2 mice. We also examined whether stem cell prepara-tions from NZB mice transplanted into DBA/2 mice couldinduce autoimmune disease.Recent approaches to stem cell separation and identifica-

tion have included separations based on density (1-4, 6),antigen or antigen-receptor expression (5, 7-10), or lectin-receptor expression (1, 2, 11-13). Using a combination ofnegative selection to deplete mature cell populations fol-lowed by positive selection of labeled stem cells, we (13) andothers (5) separated stem cells more completely than waspossible by using negative or positive selection alone. Posi-tive-selection techniques have involved labeling stem cellswith lectins--e.g., soybean agglutinin (12) or wheat germagglutinin (WGA) (1-4, 13-15)-or with monoclonal anti-bodies (mAbs) against certain stem cell surface antigens-

e.g. Thy-1 (5, 7-9) or H-2k (10)-and then using sortingtechnologies.We examined a population of low-density cells that exhibit

receptors forWGA (WGA' cells) as first described by Visserand Bol (1). WGA' cells separated from murine adult bonemarrow (WGA' BMC) are enriched for myeloid and eryth-roid progenitors that develop into late-forming (day 12)colony-forming units in spleen (CFU-S) when analyzed bythe CFU-S assay ofTill and McCulloch (16). WGA+ BMC areenriched for myeloid and erythroid progenitors that developinto mixed lineage colonies in vitro (2, 3, 17, 18). Further-more, WGA' BMC are thought to contain noncycling cells,possibly Go stem cells, because cells of this population areresistant to 5-fluorouracil treatment (4). We found thatWGA' BMC and WGA' cells isolated from fetal livercontain progenitors for lymphocytes as well as progenitorsfor cells of the myeloid and erythroid lineages (18). Lym-phocytes derived from WGA' cells developed in irradiatedmice that received WGA+ cells and were capable of respond-ing in vivo to mitogenic and antigenic stimuli specific for T orB lymphocytes (18). Earlier studies from our laboratoriesshowed that murine and simian WGA+ BMC contain im-munoregulatory (natural suppressor) activity that may beinvolved in suppression of immune function, tumor cellproliferation, and myeloid development (13, 17). Thus, itappears that WGA+ cells may indeed be important regulatorsofdevelopment and/or function of cells in the immunologicaland perhaps hematological systems.The concept that stem cell defects are of primary impor-

tance in development of systemic autoimmune disease wassupported by the findings of Morton and Siegel, who showedthat high-dose TBI of normal DBA/2 mice followed byinfusion with unseparated BMC or FLC from histocompat-ible autoimmune-prone NZB mice resulted in development ofsystemic autoimmune disease in recipients (19-22). Simi-larly, Jyonouchi et al. (23) and DeHeer and Edgington (24)demonstrated that many immunological and hematologicalabnormalities of NZB mice are transferable to lethally irra-diated DBA/2 mice. A fundamental role of stem cells in thepathogenesis of systemic autoimmune disease was supportedfurther when abnormalities characteristic of NZB mice,including development of systemic autoimmune diseases,were prevented in NZB mice that had been treated withhigh-dose TBI followed by infusion of whole BMC fromhistocompatible DBA/2 mice (23-25). More recently, Ike-hara et al. (6) initiated organ-specific, cell-specific, andsystemic autoimmune diseases by transplantation of a low-

Abbreviations: BMC, bone marrow cells; BMT, bone marrow trans-plantation; CFU-S, spleen colony-forming unit; FLC, fetal livercells; Fr-1 to Fr-5, fractions 1-5; mAb, monoclonal antibody; PLC,perinatal liver cells; TBI, total-body irradiation; WGA, wheat germagglutinin.§To whom reprint requests should be addressed.

3218

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

Apr

il 7,

202

0

Proc. Natl. Acad. Sci. USA 88 (1991) 3219

density cell population containing stem cells from autoim-mune-prone mice into autoimmune-resistant mice.Herein we report that isolated WGA' FLC or BMC from

NZB or DBA/2 mice can either produce or prevent mani-festations of systemic autoimmune disease, depending onwhether the donor strain is autoimmune-prone or autoim-mune-resistant. Low-density WGA' cells isolated fromDBA/2 mice were transferred into lethally irradiated NZBmice, and in reciprocal experiments, low-density WGA'cells separated from NZB mice were transferred into irradi-ated DBA/2 mice. Adult DBA/2 recipients of either unsep-arated, low-density, or low-density WGA' BMC or FLCfrom NZB mice regularly developed manifestations of au-toimmune disease characteristic of NZB mice. Conversely,the development of systemic autoimmune disease was pre-vented in irradiated NZB mice that received unseparated,low-density, or WGA' low-density cells separated fromDBA/2 BMC or FLC. When NZB mice were subjected toTBI and then received NZB cells i.v., the mice quicklydeveloped full-blown autoimmune disease. However,DBA/2 mice subjected to TBI and administered DBA/2 stemcell preparations i.v. did not develop autoimmune disease.These results provide additional strong evidence to implicatestem cell defects in the development of systemic autoimmunedisease in mice.

MATERIALS AND METHODSAnimals. Inbred adult 6- to 8-week-old virgin NZB/bln

mice (The Jackson Laboratory and Wadsworth Center forResearch, Albany, NY), 6- to 8-week-old virgin DBA/2 mice(The Jackson Laboratory and Charles River Breeding Lab-oratories), and fetal (gestation days 17-20) or neonatal (1-2days old) NZB/bln and DBA/2 mice obtained from ourbreeding colonies were housed in sterilized cages on laminarair flow-supplied racks and fed acidified (pH 2) water and astandard nonpurified diet ad libitum.

Preparation of Chimeric Mice. Female 19- to 21-week-oldNZB mice were lethally irradiated and then administered i.v.stem cell preparations ofBMC from 8- to 12-week-old DBA/2mice or DBA/2 FLC (gestation days 17-20). NZB mice weregiven 10.5-11.0 Gy of TBI from a 137Cs source and admin-istered i.v. one of the following: unseparated BMC or FLC(3-5 x 106); low-density, mature cell-depleted [fraction 2(Fr-2)] BMC or FLC (5 x 105 to 1 x 106); or positivelyselected, Fr-2 stem cell-enriched WGA' BMC or WGA'FLC (1 x 105).Female DBA/2 mice (8-10 weeks old) were exposed to

9.5-10 Gy of TBI and then administered i.v. stem cellpreparations from either BMC of adult NZB mice or pooledliver cells from fetal (gestation days 17-20) and neonatal (1-2days old) NZB mice, hereafter called NZB perinatal livercells (PLC).

Depletion of Mature Cells from Hematopoietic Tissues.BMC and FLC were dissociated into single-cell suspensionsand depleted of T cells, B cells, and macrophages as de-scribed (4).

Positive Selection of WGA' Low-Density BMC and FLC.BMC or FLC depleted of mature cells were fractionated byequilibrium density gradient centrifugation as described (3).Briefly, Percoll (Pharmacia) solution was prepared in densi-ties (p) of 1.090, 1.075, and 1.063 g/ml and then adjusted topH 7.0 and 300 milliosmole per kilogram of solution. Cellswere collected in fractions of various densities: p < 1.063(Fr-1), 1.063 <p < 1.075 (Fr-2), 1.075 <p < 1.090 (Fr-3), andp > 1.090 or more (Fr-4). Cells in Fr-2 were shown to beenriched for CFU-S and were kept for positive selection oflow-density, mature cell-depleted WGA' FLC and WGA'BMC as defined (3, 4, 13, 26). Cells recovered from the Fr-2gradient were adjusted to 1 x 107 cells per ml and labeled with

fluorescein isothiocyanate-conjugated WGA (Polysciences)(4). After being washed, the cells were analyzed with anEPICS C (Coulter) cell sorter with an Argon-ion, water-cooled laser with a 5-W output at 488 nm. Electronic windowswere set to include cells with low-forward (cell size) andintermediate 900 (cell granularity) light scatter. Cells that fellin the selected windows and that contained a high degree offluorescence intensity (WGA' cells) were positively selectedinto 15-ml tubes containing RPMI 1640 medium supple-mented with 10% (vol/vol) fetal bovine serum.CFU-S Assay. CFU-S were quantitated as described (13).

Briefly, 4-24 hr before cell transfer, mice were irradiatedwith 7.5-8.0 Gy and injected i.v. with appropriate concen-trations of cells. Recipient mice were sacrificed 8 and 12 daysafter transplant, and their spleens were removed and fixed inBouin's solution. Visible surface colonies were counted.Irradiated mice that did not receive cell preparations hadfewer than one colony per spleen (data not shown).

Evaluation of Serum Anti-DNA Autoantibodies. Autoanti-body titers in serum samples from mice that received trans-plants were evaluated as described (27). Serum was collectedfrom blood drawn from the retro-orbital plexus just beforesacrifice, placed in microaliquot vials (Eppendorf), and sep-arated from clotted blood by centrifugation at 16,000 x g for20 min in a micro-centrifuge (Eppendorf, 5415). Aliquots (50pl) of the separated serum were stored at -20'C until use.Calf thymus DNA (no. 2618, lot 704275; Calbiochem) wasdissolved (20 ,ug/ml) in 0.5 M carbonate buffer (pH 9.6). Onehundred microliters ofDNA solution was added to each wellof an Immulon 2, 96-well microtiter plate (Dynatech) and thenincubated for 1 hr at 370C followed by three washes withphosphate-buffered saline containing 0.05% Tween 20 (pH7.4) (washing buffer). Coated plates were stored in the darkat 40C for up to 2 weeks prior to use. Fifty microliters ofserum diluted 1:100 in washing buffer was added to each welland incubated for 1 hr at room temperature (RT). Plates werewashed three times with washing buffer, and 50 Al of a 1:350dilution of alkaline phosphatase-conjugated goat anti-mouseIg (Sigma) was added to each well. The plates were incubatedfor 1 hr at RT and then washed three times with washingbuffer. Fifty microliters of disodium p-nitrophenylphosphate(Sigma) at 1 mg/ml was added to each well, and the plate wasincubated 15-40 min at RT in the dark. Plates were read at410 nm with a Minireader 650 (Dynatech); optical densitieswere recorded and analyzed with an Apple IIC and Immu-nosoft software (Dynatech). Values were standardized byusing a sample of pooled serum from 7-month-old male NZBmice that was included in each plate of every experiment.

Histological Analysis. Histological sections (2 ,um) of thekidneys of transplanted mice and controls were preparedfrom formalin-fixed tissues and stained with periodic acid/Schiff reagent or hematoxylin and eosin stain (28). Theglomeruli were evaluated for membrane thickening and othermorphological changes and scored from 0 to 4+ (0, nodetectable pathology; 4+, gross membrane thickening indic-ative of advanced glomerular damage). A minimum of 100glomeruli from each mouse were scored blindly by twopathologists using coded samples.

Statistical Analysis. Unless otherwise stated, statisticalsignificance was determined by analysis of variance(ANOVA, Scheffe's post hoc test) and Student's t test. Pvalues < 0.05 were considered significant.

RESULTSSeparation of Low-Density WGAI Cells Enriched for He-

matopoietic Activity. Murine BMC and FLC that containedhematopoietic progenitors were separated from nonhe-matopoietic progenitors and tested for ability to form mixedlineage CFU-S as described. Flow cytometric analyses from

Immunology: Sardifia et al.

Dow

nloa

ded

by g

uest

on

Apr

il 7,

202

0

Proc. Natl. Acad. Sci. USA 88 (1991)

Table 1. Separation of WGA+ cells from murine hematopoietictissues: bone marrow (BM) and fetal liver (FL)

WGA+ cells inFr-2, % WGA+ cells

Strain (n) Tissue Fr-2, % Presort Postsort in UC, %

DBA (50) BM 4.7 17 94 0.1DBA (68) FL 2.8 26 97 0.7NZB (25) BM 2.8 23 93 0.2NZB (40) FL 6.8 12 97 0.1

Fr-2 cells are given as the percentage of unseparated BMC andFLC. The number of mice (n) is given in parentheses. UC, unsep-arated cells.

several experiments revealed that the Fr-2 cells constituted2.8-6.8% of the total unseparated cells found in murine bonemarrow or fetal liver (Table 1). WGA+ cells represented12-26% ofthe Fr-2 cells. Positive selection ofthe WGA+ cellsusing a cell sorter yielded a highly enriched, relativelyhomogeneous population ofWGA+ cells (Table 1 and Fig. 1).To determine whether enriched stem cell preparations couldbe obtained by positively selecting WGA+ cells, these cellswere compared with unseparated cells and Fr-2 cells forability to form day 8 and day 12 CFU-S. WGA+ BMC but notWGA- BMC were enriched 10- to 30-fold compared withunseparated BMC (Table 2). Fr-2 cells were also enriched incomparison with unseparated cells but were enriched lessthan WGA+ cells (18). Cells ofother densities (Fr-3) were notsignificantly enriched. Low-density WGA+ cells obtainedfrom FLC were similarly enriched for day 12 CFU-S. WGA+cells were also enriched for progenitors that differentiate intomixed lineage CFU-GEMM (granulocyte-erythrocyte-monocyte-marcrophage) (18).

Survival of Lethally Irradiated Mice That Received Trans-plants of Stem Cell Preparations. Irradiated NZB or DBA/2recipients of WGA+ cells isolated from BMC or FLC had100-day survival rates ranging from 50% for NZB recipientsof DBA/2 fetal liver to 100%o for DBA/2 recipients of NZBfetal liver (Table 3). Of 48 recipients of 1 x 105 WGA+ cells,75% survived more than 100 days. Of three DBA/2 mice thatreceived 1 x 104 WGA+ cells from NZB fetal liver, allsurvived more than 300 days (data not shown). Survivalamong WGA+ recipients was comparable to that observedamong recipients of unseparated cells (79%). Chimeric anal-ysis using fluorochrome-conjugated, allotype-specific mAbsrevealed that NZB and DBA/2 recipients ofunseparated Fr-2or WGA+ BMC or FLC were repopulated with thymocytesof donor origin (18). The data strongly suggest that WGA+BMC and FLC contain progenitors of lymphoid as well as

myeloid and erythroid cells, further supporting the conten-tion that WGA+ cells contain true hematopoietic progenitors(18). Recipients of high density (Fr-3) cells had extremelypoor 100-day survival (data not shown), as did recipients ofWGA- cells. However, NZB mice treated with 10.0-10.5 GyofTBI and then administered 5 x 105 WGA- cells i.v. had anunusually high 100-day survival rate. Chimeric analysis

-0

)t

Fluorescence intensity

FIG. 1. (Upper) Unsorted Fr-2cells (presort) contained 17.25%WGA+ cells. (Lower) The post-sort positively selected WGA+stem cell preparation contained94.47% WGA+ cells.

Table 2. Enrichment of CFU-S in low-density WGA' DBA/2 orNZB BMC

Day 8 CFU-S Day 12 CFU-S

BMC type n Mean x l0-5 SD n Mean x 10-5 SD

DBA/2UC 7 17 7 6 44 11Fr-3 5 8 3 5 9 4WGA+ 4 230 74 4 450 107WGA- 7 <1 1 7 6 4

NZBUC 6 18 4 15 28 9Fr-2 6 107 24 11 233 82Fr-3 5 36 10 5 57 10WGA+ 7 633 154 10 937 138WGA- 6 4 2 10 32 8

Mean CFU-S/105 is the arithmetic mean + SD; n is the number ofmice studied. UC, unseparated BMC cells.

showed that these mice were repopulated exclusively withthymocytes of NZB origin (18). Thus, we postulated thattransfer of a large number ofWGA- cells to NZB mice afterTBI provided the recipients with transient protection againstthe effects of TBI until their own radioresistant stem cellsrepopulated the immunologic compartment in these mice.Development of Systemic Autoimmunity in DBA/2 Recipi-

ents of NZB Stem Cell Preparations. Adult female DBA/2mice that had been administered either NZB BMC or PLCpreparations were examined 93-218 days later to determinewhether these mice developed manifestations of systemicautoimmune disease characteristic of NZB mice. DBA/2recipients of unseparated low-density or low-density posi-tively selectedWGA' cells ofNZB origin developed elevatedlevels of serum anti-DNA antibodies (Fig. 2). In addition,glomerular lesions characteristic ofNZB mice (Fig. 3 UpperLeft) that were not observed in untreated DBA/2 mice (Fig.3 Lower Left) or DBA/2 recipients of DBA/2 cells (notshown) developed in the DBA/2 mice administered unsepa-rated NZB cells (Fig. 3 Upper Right) or low-density WGA'cells (Fig. 3 Lower Left).

Prevention of Systemic Autoimmunity in NZB Recipients ofDBA/2 Stem Cell Preparations. To determine whether mar-row or stem cell replacement in autoimmune-prone miceprevents the genetically programmed development of sys-temic autoimmunity, young adult NZB mice were subjected

Table 3. Survival of lethally irradiated mice infused with NZB orDBA/2 hematopoietic cell preparations

Cell Cell dose Recipients, 100-dayCell source type x 10-5 n survival,* %

NZB -. DBA/2tBMC WGA+ 1 11 73FLC WGA+ 1 7 100BMC Fr-2 5 12 67FLC Fr-2 5 20 70BMC UC 30 10 90FLC UC 30 12 75

DBA/2 NZBtBMC WGA+ 1 19 74FLC WGA+ 1 8 50BMC Fr-2 5 8 75FLC Fr-2 10 12 100BMC UC 30 16 81FLC UC 30 5 60

Three DBA/2 recipients of 1 x 104 WGA+ NZB FLC survived >1yr after transplantation (not shown). UC, unseparated cells.*Percentage of transplanted mice surviving more than 100 days after9.5 to 10.0 Gy of TBI and stem cell transfer.

tCell donor -* recipient.

3220 Immunology: Sardifia et al.

Dow

nloa

ded

by g

uest

on

Apr

il 7,

202

0

Proc. Natl. Acad. Sci. USA 88 (1991) 3221

12 14 4 5 5 3 7 8

FIG. 2. Marked increase of anti-DNA autoantibody levels inserum in DBA/2 recipients ofNZB stem cells (N) was demonstratedby ELISA. Untreated DBA/2 (20-40 weeks old) and NZB (18-31weeks old) mice were compared with age-matched DBA/2 micestudied 100 days after transplantation with various NZB stem cellpreparations: NPL (unseparated PLC), P < 0.01; NPF (Fr-2 PLC),P < 0.001; NBW (WGA+ BMC), P < 0.001; NPW (WGA+ PLC);NBM (unseparated BMC); and NBF (Fr-2 BMC). The last threepreparations showed P < 0.001 when analyzed by Student's t test butno statistical significance when evaluated by ANOVA. The numberof mice in each group is shown.

to TBI and infused either with unseparated BMC or FLC orwith preparations of stem cell-enriched fractions as de-scribed. The NZB mice that received either Fr-2 cells or Fr-2WGA+ FLC or BMC as well as recipients of unseparatedhematopoietic cells were examined 105-152 days after trans-plantation for evidence of autoimmune disease. NZB recip-ients of low-density WGA+ cells from either DBA/2 bonemarrow or fetal liver had significantly decreased levels ofanti-DNA autoantibody levels in serum (Fig. 4). Similarly,histological examination of renal tissue from NZB recipientsof DBA/2 whole bone marrow, Fr-2 cells or low-densityWGA' stem cell preparations revealed the absence of theglomerular basement membrane thickening characteristic ofuntreated NZB mice. The glomeruli of the transplanted NZBmice (Fig. 5 Bottom) appeared normal in comparison with thestrikingly abnormal glomeruli of untreated NZB mice (Fig. 5Top) or NZB recipients of NZB low-density WGA+ BMC(data not shown).

DISCUSSIONConsiderable experimental evidence suggests that the etio-pathogenesis of systemic and organ-specific autoimmunedisease originates from defects present in hematopoietic stem

- zl -

DBA NZB DFL OFF DFW DM DBFDBW12 12 4 4 4 4 4 5

FIG. 4. Serum anti-DNA autoantibody levels were significantlyreduced in NZB recipients of DBA/2 stem cells (D). Levels mea-sured by ELISA were compared in untreated NZB (35-40 weeks old)and DBA/2 (15-34 weeks old) mice as well as in age-matched NZBmice studied >100 days after transplantation with various DBA/2stem cell preparations: DFL (unseparated FLC), P < 0.01; DFF(FLC Fr-2), P < 0.05; DFW (WGA+ FLC), P < 0.02; DBM(unseparated BMC), P < 0.02; DBF (Fr-2 BMC), P < 0.01; and DBW(WGA+ BMC), P < 0.001. The number of mice in each group isshown.

cells (20, 23, 29-32). Results of several studies conducted inour laboratories (23, 33, 34) and by others (19-22, 24, 25)indicate that replacement of stem cells using bone marrowtransplantation (BMT) can dramatically modulate develop-ment of autoimmune disease. Indeed, we have shown thatBMT can be used to treat systemic autoimmune disease inautoimmune-prone mice (23, 33, 34). Although these studiesprovided evidence that BMT may be a viable treatment forsystemic autoimmune disease, because of the difficulty ofisolating stem cell-enriched preparations, conclusive dataimplicating stem cell defects in the etiology and pathogenesisof systemic autoimmune disease was only recently obtained.

Transplantation of stem cell-enriched fractions can trans-fer systemic, cell-specific, and organ-specific autoimmunedisease from autoimmune-prone to autoimmune-resistantmice (6). Several investigators have demonstrated that mu-rine hematopoietic stem cells can be impressively concen-

FIG. 3. Periodic acid/Schiff reagent-stained kidney section froman untreated NZB mouse (Upper Left) and untreated DBA/2 mouse(LowerLeft) and from DBA/2 recipients ofunseparated PLC (UpperRight) or WGA+ PLC (Lower Right) from NZB mice. (Left andUpper Right, x 100; Lower Right, x400.)

FIG. 5. Periodic acid/Schiff reagent-stained sections of kidneyfrom an untreated NZB mouse (Top), an untreated DBA/2 mouse(Middle), and NZB recipients of DBA/2 WGA+ BMC (Bottom).(Top, x100; Middle and Bottom, x400.)

Immunology: Sardifia et al.

Dow

nloa

ded

by g

uest

on

Apr

il 7,

202

0

Proc. Natl. Acad. Sci. USA 88 (1991)

trated based on either the presence of stem cell antigens (5)or receptors for lectins (1, 2). We previously obtained stemcell-enriched fractions from murine (4, 13) and monkey (17)bone marrow by positively selecting low-density BMC thatbind WGA.

In the present analyses we found that low-density WGA'cells as well as low-density or unseparated cells obtainedfrom normal DBA/2 bone marrow or fetal liver and thentransplanted into irradiated young NZB mice greatly de-creased the levels of serum anti-DNA autoantibodies andprevented almost completely the development of glomerularlesions in recipient NZB mice. Conversely, transfer of low-density WGA' cells from NZB mice to lethally irradiatednormal DBA/2 mice resulted in development of elevatedlevels of serum anti-DNA autoantibodies and NZB-like glo-merular lesions in the DBA/2 mice. In other studies weobserved that the low-density WGA' cells contain progeni-tors for immunologically competent lymphocytes (18). SomeNZB mice treated with up to 11 Gy of TBI and then infusedwith DBA/2 WGA+ BMC or FLC possessed similar numbersof donor-derived as well as recipient-derived lymphocytes inthe spleen (18). However, in the thymuses of these mice, byfar the vast majority of lymphocytes were of donor origin.These findings suggest that despite the persistence of asignificant proportion of NZB-derived lymphocytes inspleens of recipient mice, the disease process was neverthe-less offset by stem cell transplantation, even when analyzedmore than 150 days later. Thus, cells derived from themarrow of autoimmune-resistant mice may be able to inhibitexpression of systemic autoimmune disease-a subject wor-thy of further investigation.The importance of the recipient-derived cells in the spleen

remains to be determined. However, it has been observed thatin MRL/lpr mice, the correction of autoimmune disease byusing lethal TBI followed by BMT is transient and thatmanifestations of disease are prone to recur in these animals(34). The recurrence of disease in MRL/lpr mice is correlatedwith the reappearance ofMRL/lpr-derived lymphocytes (34).Therefore, it appears that either significant improvements inmyeloablative therapy or additional restorative manipulationsto promote the persistence of the disease-correcting allogenicstem cells must be achieved to cure or prevent autoimmunityor lymphoproliferative disease that occurs in the MRL/lprmice. Recently described methods that combine myeloabla-tive drugs with TBI should now be examined to determinewhether these methods are more effective in eliminatingradioresistant progenitors from autoimmune-prone mice (35).Further studies are needed to probe the molecular and geneticnature of the stem cell defect responsible for the developmentof autoimmunity and to ascertain, if the abnormal genes areidentified, whether expression of the homologous genes fromthe autoimmune-resistant mice or the genes of the mice withautoimmune potential can be introduced appropriately toeither cause or cure the autoimmune disorder.

We thank Susan Lay and Irma Bauer-Sardifia for technical assis-tance and Jane Hardison and Ellen Lorenz for help in preparing thismanuscript.

1. Visser, J. W. M. & Bol, S. J. L. (1981) Stem Cells 1, 240-249.2. Visser, J. W. M. & de Vries, P. (1988) Blood Cells 14, 369-384.3. Lord, B. I. & Spooncer, E. (1986) Lymphokine Res. 5, 59-72.4. Miyami-Inaba, M., Ogata, H., Toki, J., Kuma, S., Sugiura, K.,

Yasumizu, R. & Ikehara, S. (1987) Biochem. Biophys. Res.Commun. 147, 687-694.

5. Spangrude, G. J., Heimfeld, S. & Weissman, I. L. (1988)Science 241, 58-62.

6. Ikehara, S., Kawamura, M., Takao, F., Inaba, M., Yasumizu,R., Than, S., Hisha, H., Sugiura, K., Koide, Y., Yoshida,T. O., Ida, T., Imura, H. & Good, R. A. (1990) Proc. Natl.Acad. Sci. USA 87, 8341-8344.

7. Berman, J. W. & Basch, R. S. (1985) Exp. Hematol. 13,1152-1156.

8. Boswell, H. S., Wade, P. M., Jr., & Quesenberry, P. J. (1984)J. Immunol. 133, 2940-2949.

9. Fredrickson, G. G. & Basch, R. S. (1989) J. Exp. Med. 169,1473-1478.

10. Mulder, A. H., Bauman, J. G. J., Visser, J. W. M., Boersma,W. J. A. & van den Engh, G. J. (1984) Cell. Immunol. 88,401-410.

11. Nicola, N. A., Burgess, W., Metcalf, D. & Battye, F. L. (1978)Aust. J. Exp. Biol. Med. Sci. 56, 663-679.

12. Reisner, Y., Itzicovitch, L., Meshorer, A. & Sharon, N. (1978)Proc. Natl. Acad. Sci. USA 75, 2933-2936.

13. Sugiura, K., Inaba, M., Ogata, H., Yasumizu, R., Inaba, K.,Good, R. A. & Ikehara, S. (1988) Proc. Natl. Acad. Sci. USA85, 4824-4826.

14. Ploemacher, R. E. & Brons, N. (1988) Exp. Hematol. 16,21-26.

15. Ploemacher, R. E. & Brons, N. (1988) Exp. Hematol. 16,27-32.

16. Till, J. E. & McCulloch, E. A. (1%1) Radiat. Res. 14,213-222.17. Sugiura, K., Ikehara, S., Gengozian, N., Inaba, M., Sardifna,

E. E., Ogata, H., Seong, M. & Good, R. A. (1990) Blood 75,1125-1131.

18. Sardifia, E. E. (1990) Dissertation (Univ. of South Florida,Tampa).

19. Morton, J. 1. & Siegel, B. V. (1974) Transplantation 17, 624-626.

20. Morton, J. I. & Siegel, B. V. (1974) Proc. Natl. Acad. Sci. USA71, 2162-2165.

21. Morton, J. I. & Siegel, B. V. (1975) Transplantation 19, 464-469.

22. Morton, J. I. & Siegel, B. V. (1978) Immunology 34, 863-868.23. Jyonouchi, H., Kincade, P. W., Good, R. A. & Fernandes, G.

(1981) J. Immunol. 127, 1232-1235.24. DeHeer, D. H. & Edgington, T. S. (1977) J. Immunol. 118,

1858-1863.25. Morton, J. I. & Siegel, B. V. (1979) Transplantation 27, 133-

134.26. Sugiura, K., Inaba, M., Ogata, H., Yasumizu, R., Sardifia,

E. E., Inaba, K., Kumu, S., Good, R. A. & Ikehara, S. (1990)Cancer Res. 50, 2582-2586.

27. Fujitsu, T., Sakuma, S., Seki, N., Senoh, H., Mori, J. &Kikuchi, H. (1986) Int. J. Immunopharmacol. 8, 897-910.

28. Clark, G. (1981) in Staining Procedures, ed. Clark, G. (Williams& Wilkins, Baltimore), 4th Ed., pp. 200-201.

29. Theofilopoulos, A. N. & Dixon, F. J. (1985) Adv. Immunol. 37,269-390.

30. Ikehara, S., Good, R. A., Nakamura, T., Sekita, K., Inoue, S.,Qo, M. M., Muso, E., Ogawa, K. & Hamashima, Y. (1985)Proc. Natl. Acad. Sci. USA 82, 2483-2487.

31. Akizuki, M., Reeves, J. P. & Steinberg, A. D. (1978) Clin.Immunol. Immunopathol. 10, 247-265.

32. Nakamura, T., Ikehara, S., Good, R. A., Inoue, S., Sekita, K.,Furukawa, F., Tanaka, H., Oo, M. M. & Hamashima, Y. (1985)Thymus 7, 151-160.

33. Himeno, K. & Good, R. A. (1988) Proc. Natl. Acad. Sci. USA85, 2235-2239.

34. Ikehara, S., Yasumizu, R., Inaba, M., Izui, S., Hayakawa, K.,Sekita, K., Toki, J., Sugiura, K., Iwai, H., Nakamura, T.,Muso, E., Hamashima, Y. & Good, R. A. (1989) Proc. Natl.Acad. Sci. USA 86, 3306-3310.

35. Lapidot, T., Terenzi, T., Singer, S., Salomon, 0. & Reisner, Y.(1989) Blood 73, 2025-2032.

3222 Immunology: Sardifia et al.

Dow

nloa

ded

by g

uest

on

Apr

il 7,

202

0