Proc. Natl. Acad. Sci. USAVol. 82, pp. 188-192, January 1985Immunology

Interferon is a mediator of hematopoietic suppression in aplasticanemia in vitro and possibly in vivo

(inhibitors/myelopoiesis/colony culture)

NICHOLAS C. ZOUMBOS*, PEDRO GASCON*, JULIE Y. DJEUt, AND NEAL S. YOUNG*

*Cell Biology Section, Clinical Hematology Branch, National Heart, Lung, and Blood Institute and tDivision of Virology, National Center for Drugs andBiologics, Bethesda, MD 20205

Communicated by Helen M. Ranney, September 20, 1984

ABSTRACT We have investigated interferon as a media-tor of hematopoietic suppression in bone marrow failure. In-terferon production by stimulated peripheral blood mononu-clear cells from patients with aplastic anemia was significantlyhigher than that observed in controls; spontaneous interferonproduction by these cells was also high for more than half ofaplastic anemia patients. Circulating interferon, not detect-able in normal individuals, was detected in 10 of 24 patients.Interferon is a potent inhibitor of hematopoietic cell prolifera-tion and, therefore, may be the mediator of suppression inmany in vitro models employing patients' cells and sera. Thepossible pathogenic importance of interferon in aplastic ane-mia was suggested by an increase in hematopoietic colony for-mation in vitro after exposure of bone marrow cells to anti-interferon antisera (277 ± 71% increase for patients comparedto 1.6 ± 1.6% for normal individuals). Interferon levels in thebone marrow sera of aplastic anemia patients were high (mean= 203 international units (IU)/ml, n = 8), even in comparisonto circulating levels in the same patients. Normal bone marrowsera also contained measurable interferon but at lower levels(41 IU/ml, n = 16), indicating that interferon may be a normalbone marrow product. High concentration of bone marrow in-terferon, possibly due to abnormal immunologic activity or areaction to virus infection of the bone marrow, may mediatehematopoietic suppression in aplastic anemia patients.

Several lines of evidence suggest that aplastic anemia maybe mediated by cells of the immune system. Occasionally,patients recover autologous hematopoietic function after im-munosuppressive therapy prior to unsuccessful bone mar-row transplantation (1, 2). Conversely, some syngeneictwins require immunosuppression to effect bone marrow en-graftment (3). Anti-thymocyte and anti-lymphocyte globulintreatments, which result in hematopoietic recovery in 40-50% of patients with aplasia, are presumed to function bytheir immunosuppressive action (4). Removal ofT cells fromaplastic bone marrow in vitro can improve hematopoieticcolony formation (5). Coculture of bone marrow or bloodcells from aplastic anemia patients with histocompatiblebone marrow from normal individuals results, in some cases,in suppression of colony formation by the normal cells (6).Hematopoietic suppressor activity also has been generatedin mixed leukocyte cultures (7).Lymphocytes that suppress hematopoiesis also may have

a role in the normal regulation of blood cell production (8). Tcells are required for erythroid colony formation from nor-mal peripheral blood in vitro, (9) a helper effect enhanced bythe prior removal of T cells bearing the HLA-DR antigen(10). Bacigalupo and his colleagues (11, 12) have generatedsuppressor cells from normal peripheral blood lymphocytesby exposing them to plant lectins; the resultant suppressor

activity is similar to that spontaneously present in cultures ofcells from patients with bone marrow failure (13).We have shown that the hematopoietic suppressor activity

in medium from lectin-stimulated T-cell cultures is y-inter-feron (14). This observation led to the current investigationof the role of interferon in hematopoietic suppression inaplastic anemia.

METHODSPatients. Twenty-three adult patients with severe aplastic

anemia, as defined by peripheral blood cell counts and bonemarrow biopsy (15), were studied before treatment with anti-thymocyte globulin or androgens. Two patients had historiesof drug or chemical exposure. Three other patients who hadachieved independence from transfusions after treatment aredesignated "post-recovery." None of the patients studiedshowed clinical evidence of active viral infection, and theirtiters of antivirus antibodies were lower than those in multi-ply transfused patients. In the control population were pa-tients with a variety of other hematologic diseases, includingmyelofibrosis, myelodysplasia, congenital pure red cell apla-sia, and chronic refractory anemia; patients with thalassemiaand sickle-cell disease who had received multiple transfu-sions; and normal, apparently healthy volunteers.

Tissues. Peripheral blood was collected in preservative-free heparin (O'Neill and Feldman, St. Louis, MO). Bloodwas allowed to clot for 1 hr before the serum was separated.T lymphocytes were separated by rosetting with sheeperythrocytes. For colony assays, bone marrow was aspirat-ed into 1 ml of Iscove's modification of Dulbecco's mediumcontaining 250 units of heparin; for interferon measure-ments, undiluted bone marrow samples were allowed to clotbefore centrifugation to separate bone marrow sera.

Interferon Measurements. Interferon was measured by itsability to inhibit cytopathic effects (CPE) of vesicular stoma-titis virus on human amnion WISH cells (16). The interferontiter is the reciprocal of the dilution that yielded 50% CPEand is expressed in international units (IU), standardizedwith National Institutes of Health reference human leuko-cyte interferon (G-023-901-527).Lymphocyte Cultures. Peripheral blood mononuclear cells

obtained by Percoll density gradient sedimentation were in-cubated for 1-7 days at 37°C in a humidified 5% CO2 atmo-sphere at a concentration of 1-2 x 106 per ml in RPMI 1640medium supplemented with 10% fetal calf serum in the pres-ence or absence of pokeweed mitogen or phytohemaggluti-nin at 10 ug/ml.

Hematopoietic Colony Culture. Mononuclear cells frombone marrow were separated by Percoll density gradientsedimentation and cultured in methylcellulose containing ap-propriate conditioned medium to support myeloid [colony-forming unit-culture (CFU-C); i.e., granulocyte/macrophage

Abbreviations: CFU-C, granulocyte/macrophage colony-formingcell; IU, international units.

188

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

June

29,

202

0

Proc. Natl. Acad Sci USA 82 (1985) 189

colony-forming cell] or erythroid [burst-forming unit-ery-throid (BFU-E); i.e., erythroid burst-colony-forming cell]colony formation (17). Interferon activity was neutralizedwith anti-a-interferon antiserum, prepared in sheep (Re-search Resources Branch, National Institute of Allergy andInfectious Diseases), anti-y-interferon antiserum, preparedin rabbits (Interferon Sciences, New Brunswick, NJ), or amonoclonal antibody to y-interferon (18) (kindly provided byB. Rubin). Controls included sheep and rabbit immune seraand monoclonal antibodies to various cell-surface determi-nants and normal human serum proteins.

RESULTSProduction of Interferon by Peripheral Blood Mononuclear

Cells in Vitro. We have shown that the hematopoietic sup-pressor activity generated by normal peripheral blood T cellsstimulated with lectins is y-interferon (14). Fig. 1 shows theresults of a representative experiment with peripheral bloodmononuclear cells from a patient with severe aplastic ane-mia; the cells were exposed to phytohemagglutinin for 7days, and the supernatant fluid was periodically sampled andtested for interferon by anti-viral assay and for hematopoiet-ic suppressor activity in normal bone marrow colony cul-ture. In comparison to normal cells (Fig. lb), cells from thepatient with aplastic anemia (Fig. la) showed a marked in-crease in total interferon produced at each of the time pointsassayed. Inhibitory activity in colony culture was more pro-nounced in the patient's cell supernatant and correlated withthe interferon concentrations in the individual supernatantsamples (Fig. 1 Inset).There were two important abnormalities in the behavior of

mononuclear cells from the aplastic anemia patients. First,the levels of interferon produced were much higher than nor-mal (Fig. 2); cells from 8 of 14 patients tested produced lev-els more than two standard deviations higher than the nor-mal mean. Cells from patients with other hematologic dis-eases and from those receiving multiple blood transfusionsdid not show abnormal production of interferon. Nor werehigh levels in the aplastic anemia patients related to the num-bers of platelet transfusions they had received; two untrans-

5000 -

4000 -

3000 -

a

50k

ED nF1 3 5 7

b 100

-E/° 10

O I/x

_ / /

0 50 100% inhibition

- 1

1 3 5 7Days of incubation

FIG. 1. Relationship of production of interferon to hematopoiet-ic inhibitor activity in peripheral blood mononuclear cell cultures.Peripheral blood mononuclear cells from a patient with aplastic ane-mia (a) or a normal individual (b) were incubated at a concentrationof 2 x 106 cells per ml with phytohemagglutinin at 10 ,Ag/ml. Super-natants were collected on days 1, 3, 5, and 7 and assayed for inter-feron by anti-viral assay (line graphs) and for hematopoietic inhibi-tor activity when present at a concentration of 10o (vol/vol) in bonemarrow tissue culture medium (bar graphs). The correlation (r =0.97) between interferon levels and inhibition of hematopoiesis inthis representative experiment is shown in the Inset.

E 3000

22000wU-

w H1000z +____---4-_ Normal

-43 5 7

DAYS OF INCUBATION

FIG. 2. Interferon production after lectin stimulation of periph-eral blood mononuclear cells. Peripheral blood mononuclear cellsfrom 4 patients with aplastic anemia and 4 hematologically normalindividuals were treated as described in the legend to Fig. 1, and theculture supernatants were assayed periodically. Both the quantity ofinterferon produced and the sustained levels at day 7 were charac-teristic of cultures of cells from aplastic anemia patients. The pro-duction of interferon in IU/ml, given as the mean ± SEM in 3-daycultures was measured for a larger population of patients and con-trols: Aplastic anemia patients, 3862 ± 1088 (n = 14); normal con-trols, 490 ± 190 (n = 16). The difference in interferon productionbetween the two populations was significant (P < 0.01).

fused patients showed marked elevations (data not shown).Second, the pattern of interferon production by the patients'cells in vitro differed from normal: in contrast to the normalfall in interferon levels by day 7, in the cultures of the pa-tients' cells interferon levels remained high throughout the 7-day incubation period (Fig. 2).

Further evidence of the disordered regulation of interferonproduction in aplastic anemia was provided by observationson unstimulated peripheral blood mononuclear cells. In theabsence of a lectin stimulant, normal cells and cells from pa-tients who had received multiple blood transfusions did notproduce detectable quantities of interferon (Table 1). In con-trast, cells from 10 of 17 patients with aplastic anemiashowed spontaneous in vitro production of interferon (16-3125 IU/ml).The inhibition of hematopoietic colony formation by su-

pernatants from lectin-stimulated cultures of patients' cellswas due to the presence of -.interferon. Addition of suffi-cient anti-y-interferon to neutralize 1000 IU of interferon tosupernatants containing comparable quantities of interferoncompletely abrogated the inhibitory effect (data not shown).This result is identical to that reported for supernatants fromlectin-stimulated normal cell cultures, in which addition ofanti-y-interferon antiserum or monoclonal antibody, and to a

Table 1. Spontaneous production of interferon by peripheralblood mononuclear cells

Interferon, No. > 10Source of cells n IU/ml IU/ml*

Aplastic anemia patients 17 213 ± 182 10/17Multiply transfused patients 10 0 0/10Normal controls 15 0.3 + 0.3 0/15

Individual values of interferon activity in culture fluid of cellsfrom aplastic anemia patients were 3125, 150, 125, 50, 50, 25, 25, 25,16, 16, 10, 5, 5, 0, 0, 0, and 0 (IU/ml). Only one normal person'scells produced detectable interferon activity of 5 IU/ml. The levelsfor 14 other normals and for 10 multiply transfused patients were 0IU/ml.*P < 0.001 for the differences between the aplastic anemia groupand each of the other two groups.

2000

1000

50

EC~20a)

r-

0

C0

0C) 10

Immunology: Zournbos et aL

Dow

nloa

ded

by g

uest

on

June

29,

202

0

190 Immunology: Zoumbos etaLP

Table 2. Circulating interferon levels

Interferon, No. > 10n IU/ml IU/ml

Aplastic anemia patients 24 87 ± 30 10/24Normal persons 16 <10 0/16Multiply transfused pa-

tients with hereditarydiseases 18 <10 0/18

lesser degree anti-a-interferon antiserum, reversed the in-hibitory effect (14).Serum Interferon Levels. Circulating levels of interferon

were measured in aplastic anemia patients, patients who hadreceived multiple transfusions, and normal controls (Table2). In both the normal and the multiply transfused controlpopulations, interferon was not detectable in sera (5 IU/mlwas the limit of sensitivity of the virus protection assay).However, in 42% of the patients with aplastic anemia, highlevels of interferon were measured. There was no correla-tion between high serum levels of interferon and spontane-ous production of interferon by peripheral blood mononucle-ar cells in vitro.

Effect of Anti-Interferon on Bone Marrow Colony Forma-tion in Vitro. We tested the ability of bone marrow mononu-clear cells to form hematopoietic colonies in vitro in the pres-ence of anti-interferon antisera. In cultures of 10 normalbone marrows, there was no effect on colony formation inthe presence of anti-y-interferon sufficient to neutralize~800 IU of interferon per 1-ml dish; in 4 of the normal cul-tures, anti-a-interferon at 5000 neutralizing units per dishwas also without effect. In contrast, anti--interferon antise-rum significantly increased hematopoietic colony formationby CFU-C from 9 of 10 patients and returned colony numberto the normal range in 4 cases (Fig. 3). Similar results wereobtained for BFU-E (erythroid)-derived colonies (data not

CU)-J-Jw

c:

w-J

z0z0U,

C-i

LOR0

cU

z0L

U

shown). CFU-GEMM (granulocyte/erythroid/macrophage/megakaryocyte colony-forming cell)-derived colonies werenot observed in control or anti-interferon-treated dishes. Incultures from two patients tested, a monoclonal antibody to-interferon was as effective as the anti--interferon antise-

rum in increasing colony formation (data not shown). Anti-a-interferon antiserum showed a similar but less dramatic ef-fect on colony formation. Manipulations of culture condi-tions that served as controls for the antisera effects includedincubation with a sheep antiserum against an interferon-defi-cient antigen and hyperimmune rabbit sera from multiplesources. An average 4-fold increase in colony formation withanti-y-interferon was observed in the cultures from theaplastic patients (Table 3), although the interpretation of theincreases must be tempered by the finding of very low colo-ny number in untreated dishes in some cases.

Interferon in Aplastic Bone Marrow. Interferon was mea-sured in bone marrow sera in the last patients studied in thisseries (Table 4). All 8 patients with aplastic anemia had inter-feron in their bone marrow sera at concentrations that resultin significant inhibition of hematopoietic colony formation invitro (14, 19). Interferon activity also was present in normalbone marrow sera but at significantly lower levels than thatmeasured in bone marrow from aplastic anemia patients.Mononuclear bone marrow cells from all 6 patients testedshowed significantly increased colony formation in vitro inthe presence of anti-interferon antiserum. Bone marrow seralevels of interferon were higher than simultaneously ob-tained circulating sera levels in 6 patients tested, suggestinglocal production or concentration of interferon in the bonemarrow.

DISCUSSIONVirtually every patient with aplastic anemia reported in thisstudy showed abnormal interferon production in vitro. Whenexposed to lectins, the mononuclear cells from their periph-

50

40

30

2010

260

50

40

30

20

10

cd e3 4 5

a b c d e a b c d e a b d e a b c d e a c e6 7 8 9 10

FIG. 3. Effect of anti-interferon antisera on hematopoietic colony formation in vitro by bone marrow cells from patients with aplasticanemia. Bone marrow mononuclear cells from 10 untreated patients with severe aplastic anemia were cultured in methylcellulose at 1 x 10'cells per ml and CFU-C-derived colonies were scored on day 10. Each group of bars represents the results obtained with cells from one of thepatients. The culture medium contained no antiserum (bars a), control hyperimmune sheep antiserum (bars b), control hyperimmune rabbitantiserum (bars c), sheep antiserum to human a-interferon (5000 neutralizing units/ml) (bars d), or rabbit antiserum to human yinterferon (800neutralizing units/ml) (bars e). Error bars represent standard deviations of replicate plates. Significant differences (by Student's t test) betweenplates containing anti-y~interferon and control plates were observed for patients 1, 2, and 6 (P < 0.001); 3, 4, 5, 7, and 9 (P < 0.01); and 10 (P <0.02). The number of bone marrow CFU-C per 105 mononuclear cells in cultures from 10 normal persons was 91 ± 32 (mean ± SD) and wasunaffected by anti-y-interferon.

Proc. NatL Acad Sd USA 82 (1985)

1

=em ffi ri

Dow

nloa

ded

by g

uest

on

June

29,

202

0

Proc. NatL Acad Sci USA 82 (1985) 191

Table 3. The effect of anti-interferon antisera on hematopoieticcolony formation (CFU-C)

% increase*Source of bone marrow cells Anti-a- Anti-y-

Groupt Condition n interferon interferon

A Aplastic anemia, prior 10 158 ± 37 277 ± 71to therapy

B Aplastic anemia, 3 14 ± 5 54 ± 29post-recovery

C Other hematologic 12 32 ± 31 10 ± 10diseases

D Normal 10 -0.5 ± 2 1.6 ± 1.6

*Relative to control cultures from the same individuals. Given asmean ± SEM.

tStatistical analysis of differences in anti-y-interferon effect: be-tween A and D, P < 0.001; A and C, P < 0.001. Anti-a-interferoneffect: between A and D, P < 0.02; A and C, P < 0.05.

eral blood showed very high levels of interferon productionin comparison to cells from either normal persons or thosereceiving blood transfusions for other hematologic diseases.Furthermore, interferon production in vitro was sustained,with elevated amounts of interferon present in the cell super-natants 7-8 days after initiation of cultures, at which timenormal cell culture interferon concentrations had returnedto base-line values. Interferon is not normally measurable incultures of lymphocytes unstimulated by lectins, butin cultures from about half of the aplastic anemia patientsinterferon was produced spontaneously, in some cases atvery high concentrations. Interferon production by cellsfrom these patients correlated with activity that suppressednormal hematopoietic colony formation in vitro, an effectcompletely reversed by the prior addition of antisera to yinterferon.The identification of a specific lymphokine as a hemato-

poietic inhibitor has implications for the interpretation ofpreviously published experiments in which in vitro systemswere used to demonstrate suppression in aplastic anemia. Inretrospect, many experiments showing hematopoietic sup-pression in vitro for cells from patients with aplastic anemialikely were measuring the effects of interferon. Coculture ofnormal and aplastic anemia bone marrow or blood cells (6,20), mixed leukocyte cultures (7), and lectin stimulation ofmononuclear cells (11) are all conditions that promote inter-feron, particularly y-interferon, production (21, 22). In otherexperiments, increased colony formation by aplastic bone

Table 4. Interferon activity in bone marrow serum andblood serum

Interferon, IU/mlBone marrow Blood

Aplastic anemia patient3 250 504 227 906 250 ND7 153 309 125 ND

11 153 1012 270 9013 200 50

Mean ± SD 203 ± 54* 53 ± 32Normal individuals (n = 16)

Mean ± SD 41 ± 23* <10

marrow has been observed after removal of T lymphocytes(5). Interferon may also be implicated in these experiments,as T cells are the source of -interferon (23) and also may berequired to mediate its effects upon target cells (24).Does interferon play a role in the suppression ofbone mar-

row function in patients with aplastic anemia, or are the ab-normalities described epiphenomena associated with bonemarrow failure and its treatment? Many patients with aplasiashowed evidence of circulating interferon. These observa-tions were made in patients who, in addition to having littleabnormal susceptibility to viral disease, showed no clinicalevidence of active viral infection. Serum interferon levels, aswell as in vitro production of interferon, were not abnormalin a control population of multiply transfused patients whohad been more frequently exposed to the same transfusion-transmitted infections and who had higher levels of serumantibodies to hepatitis viruses, Epstein-Barr virus, and cyto-megalovirus (25). Some of the patients with aplastic anemiawith marked abnormalities of interferon production hadtransfusion histories limited to only a few erythrocyte units.Our results are also consistent with the observation of bonemarrow toxicity in cancer patients treated with intravenousinfusions of recombinant interferon (19).

Levels of interferon were elevated in the bone marrowsera of patients with aplastic anemia, and the concentrationsmeasured were in the range known to inhibit hematopoieticcell proliferation in vitro (14, 26). In addition, colony forma-tion by aplastic bone marrow was improved by antisera tointerferon, suggesting that hematopoietic suppression in thepatients might be the result of the presence of interferon.Interferon activity was also present in sera from normal bonemarrow, but at a significantly lower level than in aplasticbone marrow. Subtyping of the interferons present in aplas-tic bone marrow sera has shown the presence of both y- anda-interferon (unpublished observations). Synergy in the anti-proliferative effects of y- and a-interferon has been demon-strated with murine bone marrow (27) and leukemic cell lines(28). The higher levels of interferon in the bone marrow com-

pared to the blood in aplastic anemia patients suggest thatmarrow is the origin of interferon.Bone marrow interferon also was present, although at low-

er concentrations, in normal individuals, in whom interferonis not detectable in sera from circulating blood. Only a-inter-feron, and not y-interferon, has been detected in subtypingof normal bone marrow sera (unpublished observations).That interferon might be a normal bone marrow product alsohas been suggested by the measurement of high levels ofinterferon in cultures of human marrow stromal cells (29). Incontrast to its effect on aplastic bone marrow, addition ofanti-interferon antisera to normal bone marrow cultures didnot alter colony formation. The levels of interferon presentin normal bone marrow or produced during normal marrowcell culture may not inhibit cell proliferation. Studies withmouse bone marrow cells suggest that preincubation with in-terferon (at low concentrations similar to those that we de-tected in normal human bone marrow) may enhance myeloidcell sensitivity to growth factors (30). Interferon first wasdefined by its ability to protect cells from viral infections(31); its roles in cell-cell interactions (32) and suppression ofcell proliferation were recognized subsequently (33), but apossible regulatory function in hematopoiesis suggests thatthe description of interferon as an "immune mediator" maybe too restrictive.Why is interferon production abnormal in aplastic anemia?

Abnormalities of interferon production may be the conse-quence of the grossly disordered bone marrow that charac-terizes aplastic anemia; dysregulation of the production ofother lymphokines, such as interleukins 1 and 2, is also com-mon in aplastic anemia patients (34). yInterferon productionmay be the result of more specific inciting events. For exam-

Patient numbers 3-9 refer to Fig. 3. Patient 11 was partially re-covered after anti-thymocyte globulin treatment. Patients 12 and 13had been untreated.*P < 0.001.

Immunology: Zournbos et aL

Dow

nloa

ded

by g

uest

on

June

29,

202

0

192 Immunology: Zoumbos et aL

ple, sensitized T cells produce -interferon after reexposureto virus (35), and the recent finding that a form of aplasia iscaused by a parvovirus (36) raises the possibility that otherforms of bone marrow failure might result from viral infec-tion and perhaps be mediated by interferon's suppressive ef-fects (37). Interferon-producing lymphocytes, whether aclone of abnormal suppressor cells or abnormally regulatednormal cells, may be the targets of anti-thymocyte globulinand anti-lymphocyte globulin therapy (2).

We thank Drs. Arthur Nienhuis and Eric Raefsky for advice andcareful reading of the manuscript and thank Imogene Surrey for herhelp in preparation of the paper.

1. Mathe, G., Amiel, J. L., Schwarzenberg, L., Choag, J., Tro-lard, P., Schneider, M., Hagat, M., Schlumberger, J. R. & Jas-min, C. (1970) Br. Med. J. 2, 131-136.

2. Young, N. (1981) Prog. Hematol. 12, 227-274.3. Appelbaum, F. R., Fefer, A., Cheever, M. A., Sanders, J. E.,

Singer, J. W., Adamson, J. W., Mickelson, E. M., Hansen,J. A., Greenberg, P. P. & Thomas, E. D. (1980) Blood 55,1033-1039.

4. Young, N. & Speck, B. (1984) in Aplastic Anemia: Stem CellBiology and Advances in Treatment, eds. Young, N., Levine,A. & Humphries, R. K. (Liss, New York), pp. 221-226.

5. Torok-Storb, B., Sieff, C., Storb, R., Adamson, J. & Thomas,E. D. (1980) Blood 55, 211-215.

6. Singer, J., Doney, K. & Thomas, E. D. (1979) Blood 54, 180-185.

7. Torok-Storb, B. (1984) in Aplastic Anemia: Stem Cell Biologyand Advances in Treatment, eds. Young, N., Levine, A. &Humphries, R. K. (Liss, New York), pp. 163-172.

8. Zoumbos, N., Gascon, P. & Young, N. (1983) Blut 48, 1-9.9. Nathan, D., Chess, L., Hillman, D., Clarke, B., Breard, J.,

Merler, G. & Housman, D. (1978) J. Exp. Med. 147, 324-339.10. Torok-Storb, B. & Hansen, J. A. (1982) Nature (London) 298,

473-474.11. Bacigalupo, A., Podesta, M., Mingari, M., Moretta, L., Piag-

gio, G., Van Lint, M., Durando, A. & Marmont, A. (1981)Blood 57, 491-496.

12. Podesta, M., Frassoni, F., Van Lint, M., Piaggio, G., Mar-mont, A. & Bacigalupo, A. (1982) Exp. Hematol. 10, 256-262.

13. Bacigalupo, A., Podesta, M., Frassoni, F., Piaggio, G., VanLint, M., Raffo, M., Repetto, M. & Marmont, A. (1982) Br. J.Haematol. 52, 421-427.

14. Zoumbos, N., Djeu, J. & Young, N. (1984) J. Immunol. 133,769-774.

15. Camitta, B. M., Thomas, E. D., Nathan, D. G., Santos, G.,Gordon-Smith, E. C., Gale, R. D., Rappeport, J. M. & Storb,R. (1976) Blood 48, 63-70.

16. Djeu, J., Stocks, N., Zoon, K., Stanton, G., Timonen, T. &Herberman, R. (1982) J. Exp. Med. 156, 1222-1234.

17. Iscove, N., Senn, J., Till, J. & McCulloch, E. (1971) Blood 37,1-5.

18. Rubin, B., Bartal, A., Anderson, S., Millet, S., Hirshaut, Y. &Feit, C. (1983) J. Immunol. 130, 1019-1020.

19. Gutterman, J., Fine, S., Quesada, J., Horning, S. J., Levine,J. F., Alexanian, R., Bernhardt, L., Kramer, M., Spiegel, H.,Colburn, W., Trown, P., Merigan, T. & Dziewanowski, Z.(1982) Ann. Intern. Med. 96, 549-556.

20. Kagan, W., Ascensao, J., Pahwa, R., Hansen, J., Goldstein,G. & Valera, L. B. (1976) Proc. Natl. Acad. Sci. USA 73,2890-2894.

21. Wheelock, E. (1965) Science 149, 310-311.22. Gifford, G., Tibor, A. & Peavy, D. (1971) Infect. Immun. 3,

164-169.23. Kasahara, T., Hooks, J., Dougherty, S. & Oppenheim, J.

(1983) J. Immunol. 130, 1784-1789.24. Aune, T. & Pierce, C. (1982) Proc. Nati. Acad. Sci. USA 79,

3808-3812.25. Gascon, P., Zoumbos, N. & Young, N. (1984) Ann. Intern.

Med. 100, 173-177.26. Broxmeyer, H., Lu, L., Platrer, E., Feit, C., Juliano, L. &

Rubin, B. (1983) J. Immunol. 131, 1300-1305.27. Klimpel, G., Fleischmann, W. R., Jr., & Klimpel, K. (1982) J.

Immunol. 129, 76-80.28. Fleischmann, W. R., Jr. (1982) Cancer Res. 42, 869-875.29. Shah, G., Dexter, T. M. & Lanotte, M. (1983) Br. J. Haema-

tol. 54, 365-372.30. Moore, R. & Rouse, B. (1983) J. Immunol. 131, 2374-2378.31. Isaacs, A. & Lindenmann, J. (1957) Prog. R. Soc. London Ser.

B. 147, 258-267.32. Gresser, I. (1977) Cell. Immunol. 34, 406-415.33. Blalock, J., Georgiades, J., Langford, M. & Johnson, H. (1980)

Cell. Immunol. 49, 390-394.34. Gascon, P., Zoumbos, N., Scala, G., Moore, J. & Young, N.

(1985) Blood, in press.35. Stanton, G., Langford, M. & Weigent, D. (1982) Tex. Rep.

Biol. Med. 41, 84-88.36. Young, N., Mortimer, P., Moore, J. & Humphries, R. K.

(1984) J. Clin. Invest. 73, 224-230.37. Young, N. & Mortimer, P. (1984) Blood 63, 729-737.

Proc. NatL Acad ScL USA 82 (1985)

Dow

nloa

ded

by g

uest

on

June

29,

202

0